blindoracle MCP server
io.github.craigmbrown/blindoraclecraigmbrownstreamable-httpv2.0.5Scanned
Trust layer for x402 A2A trades: reputation, pre-hire checks, escrow, disputes, delegation chains
55 tools: 53 change things and 2 only read.
- Reads private data · 1
- Reads text others write · 53
- Can send data out · 0
- 4 findings
Tools listed without sign-in. Remote posture
With Anlyon
53 tools here change things. Govern them.
- Anlyon lists the tools and adopts none.
- Each tool you adopt gets approvals and receipts.
- Your agent never holds the key.
A tool your agent calls directly is not routed through Anlyon and is not governed by it. Anlyon Radar only lists and labels, and these labels are the same either way.
What the blindoracle MCP server can do
The blindoracle MCP server is published by the GitHub account craigmbrown. It has 55 tools. It can create or change data (53 tools: health_check, agent.prehire-check, agent.trust-badge, arbitration.dispute-settlement and 49 more), but cannot delete data. 2 tools only read.
1 tool reads private data (list_services); 53 tools bring in text other people can write (agent.prehire-check, agent.trust-badge, arbitration.dispute-settlement and 50 more).
The fixed checks matched 4 findings (credentials-mention and long-description). Each one comes with the text it matched.
Its remote endpoint lists its tools without sign-in, and its tools changed 3 times since Anlyon Radar first saw them.
Last scanned Nov 19, 2026, version 2.0.5.
Tools
Grouped by what each tool does, most consequential first. Open one for its reasons and full description; descriptions are the publisher's own words.
55 tools
Changes things 53Claude Code: ask
agent.prehire-checkPre-hire due-diligence check on an agent before you delegate it real spend authority: settlement history, dispute rate, and any flagged incidents, in one signed report.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Pre-hire due-diligence check on an agent before you delegate it real spend authority: settlement history, dispute rate, and any flagged incidents, in one signed report. (x402: $0.25 USDC per call)
Annotations the server declares
- title: Pre-Hire Agent Check
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
agent.trust-badgeFREE over MCP: 5 calls per day in total, one pool shared by agent_trust-badge and reputation_lookup; no wallet, no API key.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it FREE over MCP: 5 calls per day in total, one pool shared by agent_trust-badge and reputation_lookup; no wallet, no API key. A lookup that finds nothing does not use one. Free only when the call carries NO payment: a call that sends a payment or starter-credit note is charged. queryable trust badge for an agent holding a BlindOracle passport: AUDITED when an in-date BlindOracle security audit of its code is on record (with risk score, proof id and expiry), otherwise identity only. An agent with no record returns UNVERIFIED, never a synthetic score. After the free calls: $0.01 USDC via x402 (network eip155:8453).
Annotations the server declares
- title: Agent Trust Badge
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
arbitration.dispute-settlementAdjudication of a contested deliverable: both sides submit evidence and a signed verdict (upheld / overturned / withdrawn; ProofOfAdjudicatedOutcome 30129) is written to the settle…Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Adjudication of a contested deliverable: both sides submit evidence and a signed verdict (upheld / overturned / withdrawn; ProofOfAdjudicatedOutcome 30129) is written to the settlement ledger. Disclosure: today the adjudicator is the BlindOracle operator panel — the verdict is unilateral and there is no buyer contest channel yet. It is a documented, evidence-retained decision, not a disinterested third-party referee; the live /v1/services `dispute_disclosure` block is authoritative. No charge if the required source cannot be fetched. (x402: $5.00 USDC per call)
Annotations the server declares
- title: Dispute Settlement — Operator-Panel Adjudication (disclosed)
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
attestation.single-use-sealA single-use cryptographic seal bound to an attestation or proof id: the holder can present it exactly once, and a second presentation is refused (ProofOfSealRedemption 30119).Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it A single-use cryptographic seal bound to an attestation or proof id: the holder can present it exactly once, and a second presentation is refused (ProofOfSealRedemption 30119). Called with a delegation_token and deliverable_sha256 from a recorded delegation, it instead seals that deliverable hash into the delegation chain (AgentDeliverySeal). (x402: $0.05 USDC per call)
Annotations the server declares
- title: Single-Use Attestation Seal
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
bundle.delegation-chainOne payment for a two-party delegation record: you record the delegation now (signed, kind 30014, bound to your paying wallet), and the delegate's accept, the delivery seal and you…Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it One payment for a two-party delegation record: you record the delegation now (signed, kind 30014, bound to your paying wallet), and the delegate's accept, the delivery seal and your receipt close are prepaid. Each later step is redeemed with its party's one-time token and costs nothing more. Steps not redeemed by the expiry (7 days) are refunded on the rail you paid with. Verify the whole chain at /v1/delegations/{id} with craigmbrown.com's DID document. (x402: $0.16 USDC per call)
Annotations the server declares
- title: Proof of Delegation — Prepaid Chain
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
bundle.dispute-evidencePull the respondent's track record and trust record as evidence, run the dispute settlement, then seal the verdict with a one-time seal.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Pull the respondent's track record and trust record as evidence, run the dispute settlement, then seal the verdict with a one-time seal. (x402: $4.56 USDC per call)
Annotations the server declares
- title: Dispute With Evidence
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
bundle.due-diligenceConfirm the entity exists in a public registry, pull any BlindOracle trust record, then run the $1 due-diligence scan.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Confirm the entity exists in a public registry, pull any BlindOracle trust record, then run the $1 due-diligence scan. Stops if the registry page does not name the entity. (x402: $1.00 USDC per call)
Annotations the server declares
- title: Company Due-Diligence Ladder
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
bundle.enterprise-ladderSingle-agent MASSAT audit and an injection test first; the $25 enterprise audit runs only if neither finds a blocking problem, then the result is notarized.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Single-agent MASSAT audit and an injection test first; the $25 enterprise audit runs only if neither finds a blocking problem, then the result is notarized. Process attestation runs only if you supply run evidence. (x402: $28.12 USDC per call)
Annotations the server declares
- title: Enterprise Audit Ladder
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
bundle.massat-ladderTrack record ($0.01) then a pre-hire background check ($0.25); the $5 MASSAT audit runs only if the pre-hire verdict is not AVOID.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Track record ($0.01) then a pre-hire background check ($0.25); the $5 MASSAT audit runs only if the pre-hire verdict is not AVOID. One payment; unexecuted steps refunded. (x402: $4.73 USDC per call)
Annotations the server declares
- title: Agent Security Ladder
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
bundle.prehire-ladderTrack record plus pre-hire check; the $5 technical audit runs only when you set high_stakes:true and give a target, and never after an AVOID verdict.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Track record plus pre-hire check; the $5 technical audit runs only when you set high_stakes:true and give a target, and never after an AVOID verdict. (x402: $4.73 USDC per call)
Annotations the server declares
- title: Pre-Hire Ladder
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
bundle.reputation-badgeLook up an agent's track record, then publish it as a queryable trust badge.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Look up an agent's track record, then publish it as a queryable trust badge. Stops if the agent is not found. (x402: $0.02 USDC per call)
Annotations the server declares
- title: Reputation + Badge
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
bundle.researched-debateGather sourced facts first, then run the multi-agent debate over them.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Gather sourced facts first, then run the multi-agent debate over them. Stops before the debate if the research found too little signal. (x402: $1.84 USDC per call)
Annotations the server declares
- title: Researched Debate
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
bundle.seal-and-badgeVerify who you are stamping for, issue a one-time seal, then publish the trust badge.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Verify who you are stamping for, issue a one-time seal, then publish the trust badge. Stops if the agent is unknown or has revocations. (x402: $0.06 USDC per call)
Annotations the server declares
- title: Verified Seal + Badge
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
bundle.smb-agent-guardrailsFor a business putting an agent with its own email or phone number in front of customers.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it For a business putting an agent with its own email or phone number in front of customers. Track record and pre-hire check on the agent, a prompt-injection scan of its system prompt and tools when you send them, then a signed delegation record naming the actions it must get a human to confirm first (refunds, outbound email, ...). Stops before the record on an AVOID verdict or a flagged scan. BlindOracle signs and publishes the policy; it does not sit in the agent's phone or mail path and does not enforce it. Attach a person grant to cap spend and revoke it in one call. One payment; unexecuted steps refunded. (x402: $0.77 USDC per call)
Annotations the server declares
- title: Before Your Agent Gets a Phone Number
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
bundle.trust-paperworkConfirm the entity is in a public registry, then pull its BlindOracle trust record.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Confirm the entity is in a public registry, then pull its BlindOracle trust record. Stops if the registry page does not name it. (x402: $0.10 USDC per call)
Annotations the server declares
- title: Entity + Trust Record
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
bundle.vendor-vettingVendor onboarding in one payment: registry check, trust record and the $1 DD scan, plus an optional wallet sanctions screen and a near-duplicate check against your vendor list.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Vendor onboarding in one payment: registry check, trust record and the $1 DD scan, plus an optional wallet sanctions screen and a near-duplicate check against your vendor list. Returns onboarding.decision (CLEAR | EXCEPTION | STOP) and only the exceptions a human must resolve, with a per-step baseline. The $99 deep vetting is opt-in (skip_deep_vetting=false). Stops if the registry page does not name the vendor. Unexecuted steps are refunded. (x402: $90.10 USDC per call)
Annotations the server declares
- title: Vendor Onboarding — one decision, only the exceptions
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
content.youtube-researchResearches YouTube videos from their real transcripts: pass video URLs in the task or in `urls`, or describe a topic and it finds relevant videos.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Researches YouTube videos from their real transcripts: pass video URLs in the task or in `urls`, or describe a topic and it finds relevant videos. Returns what you asked for, then per-video thesis, key claims and quotes, each cited to a transcript timestamp [n @ mm:ss]. If no transcript can be fetched, the job is refused and you are not charged. Refuses, with no charge, when fetched transcript cannot be fetched. (x402: $0.03 USDC per call)
Annotations the server declares
- title: YouTube Transcript Research
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
crypto.market-analyzerReal-time market data, technical indicators, and sentiment for any ticker, run by crypto-market-agent-sonnet against live exchange data rather than cached snapshots.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Real-time market data, technical indicators, and sentiment for any ticker, run by crypto-market-agent-sonnet against live exchange data rather than cached snapshots. Settlement proof: ProofOfSettledOutcome (kind 30120, data/proof_settled_outcomes.jsonl). Broader single-ticker view than any one oracle.* SKU. May deliver with a first-line DEGRADED notice when live data is unavailable. (x402: $0.02 USDC per call)
Annotations the server declares
- title: Crypto Market Intelligence
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
data.business-registryPublic business-registry record extraction (SEC / state Secretary-of-State / UK Companies House) over a buyer-supplied public records URL, wrapped in the BlindOracle trust envelope…Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Public business-registry record extraction (SEC / state Secretary-of-State / UK Companies House) over a buyer-supplied public records URL, wrapped in the BlindOracle trust envelope. Public records only. (x402: $0.10 USDC per call)
Annotations the server declares
- title: Business Registry Lookup
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
data.sec-edgar-filingPer-call retrieval of recent SEC EDGAR filings (10-K/10-Q/8-K) for a ticker or CIK, with a tamper-evident BlindOracle trust envelope (content hash + content-trap scan + provenance)…Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Per-call retrieval of recent SEC EDGAR filings (10-K/10-Q/8-K) for a ticker or CIK, with a tamper-evident BlindOracle trust envelope (content hash + content-trap scan + provenance). Public-domain source. (x402: $0.05 USDC per call)
Annotations the server declares
- title: SEC EDGAR Filing Retrieval
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
data.web-extractClean main-content extraction of a single buyer-supplied URL via Firecrawl, wrapped in the BlindOracle trust envelope.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Clean main-content extraction of a single buyer-supplied URL via Firecrawl, wrapped in the BlindOracle trust envelope. Thin extraction tool; no data inventory held. (x402: $0.05 USDC per call)
Annotations the server declares
- title: Clean Web Extract (per URL)
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
delegation.acceptThe delegate's countersignature on a recorded delegation: send the delegate token from delegation.record, and optionally an EIP-191 signature from the delegate wallet named in the …Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it The delegate's countersignature on a recorded delegation: send the delegate token from delegation.record, and optionally an EIP-191 signature from the delegate wallet named in the record (acceptanceStrength 'wallet' instead of 'token'). Returns a signed AgentDelegationAcceptance that hashes the record. Free when the delegation was bought as bundle.delegation-chain. (x402: $0.02 USDC per call)
Annotations the server declares
- title: Proof of Delegation — Accept
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
delegation.recordRecords that the paying agent delegated a task to a named agent: a signed credential (did:web:craigmbrown.com#delegation-1, kind 30014) binding the paying wallet, the delegate, the…Reads text others write1 finding
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Findings
- Info
The description mentions "credential".
What this check looks forMatched text
description: "...gated a task to a named agent: a signed credential (did:web:craigmbrown.com#delegation-1,..."
Description, as the server wrote it Records that the paying agent delegated a task to a named agent: a signed credential (did:web:craigmbrown.com#delegation-1, kind 30014) binding the paying wallet, the delegate, the sha256 of the task, scope, deadline and an optional policy naming the actions the delegate must get a human to confirm first. Returns two one-time tokens, one per party, for the later steps: delegate accept, delivery seal, delegator close. Anyone can verify the chain with craigmbrown.com's DID document. It proves who paid and what was handed over, not either party's off-platform identity or the quality of the work. (x402: $0.10 USDC per call)
Annotations the server declares
- title: Proof of Delegation — Record
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
deliberation.multi-agent-debate5-11 agent panel debate with 11 LLM models, structured voting, forced decision-making May deliver with a first-line DEGRADED notice when buyer question is unavailable.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it 5-11 agent panel debate with 11 LLM models, structured voting, forced decision-making May deliver with a first-line DEGRADED notice when buyer question is unavailable. (x402: $2.00 USDC per call)
Annotations the server declares
- title: Multi-Agent Deliberation Council
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
finops.token-spend-auditIndependent audit of an agent's actual token spend against its budget and declared task scope — surfaces cost overruns or scope creep a counterparty wouldn't otherwise see.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Independent audit of an agent's actual token spend against its budget and declared task scope — surfaces cost overruns or scope creep a counterparty wouldn't otherwise see. (x402: $0.05 USDC per call)
Annotations the server declares
- title: Token Spend Audit
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
health_checkCheck BlindOracle service health and API statusNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it Check BlindOracle service health and API status
Annotations the server declares
- title: Health check
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
ops.agent-workforce-diagnostic$0.25 entry rung of the Agent Workforce Early-Warning Report: agent:human hours (40 h/week parity + measured engaged) and human-wait share with the > 30 min wait exclusion, compute…Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it $0.25 entry rung of the Agent Workforce Early-Warning Report: agent:human hours (40 h/week parity + measured engaged) and human-wait share with the > 30 min wait exclusion, computed deterministically over YOUR OWN Claude Code transcript bundle with the method of Anthropic's ‘How employment is changing at Anthropic’ (2026-10-16). Zero LLM, zero network; only the bundle's sha256 is retained. Upgrade to ops.agent-workforce-report ($25) for reading load, meta-work share, skill use and the full parity notes. Bundle cap: 4 MB per request on this endpoint. Delivered as a written Markdown report (`output`) beside the `metrics` JSON; the optional git_log / decisions / instructions inputs of the $25 report are ignored at this tier. Sending more than 4 MB of transcripts, or do not want to send them at all? Run the free local extractor https://craigmbrown.com/blindoracle/tools/workforce_extract.py (sha256 34f33389a564555e4d4476b86982feff10ca53eb6143bbc38139798d8dd09417) and send its `summary` instead of `records`: counts only, no message text, any number of days, identical numbers. The report then states that it was computed from a buyer-run extraction and whether your extractor matches the published hash; the attestation proves that input produced this output, not that the input is complete. (x402: $0.25 USDC per call)
Annotations the server declares
- title: Agent Workforce Diagnostic (hours + wait share)
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
ops.agent-workforce-reportDeterministic Agent Workforce Early-Warning Report over YOUR OWN Claude Code transcript bundle, computed with the method of Anthropic's ‘How employment is changing at Anthropic’ (2…Reads text others write1 finding
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Findings
- Info
The description is 2,079 characters long, over the 2,000 this check looks for.
What this check looks forMatched text
description ends: "...roduced this output, not that the input is complete. (x402: $25.00 USDC per call)"
Description, as the server wrote it Deterministic Agent Workforce Early-Warning Report over YOUR OWN Claude Code transcript bundle, computed with the method of Anthropic's ‘How employment is changing at Anthropic’ (2026-10-16). Delivered as a written Markdown report (`output`, also at /v1/services/result/{job_id}?format=md) with one row per metric: your value | Anthropic's figure | what it means, plus the structured `metrics` JSON. Metrics: agent:human hours (40 h/week parity, pro-rated to the observed span when the bundle is short, and measured engaged time), human-wait share with the > 30 min wait exclusion, reading load (words/day at 250 wpm), meta-work share (deterministic path rule), skill use; and, when you also send them, R6 code survival by author class from `git_log`, decision latency from `decisions`, and the R5 standing-instruction inventory from `instructions`. Anything not supplied is listed with the export command. Zero LLM, zero network; inputs are processed in memory and only their sha256 is retained — no message text, paths, shas, subjects or ids are echoed. The full report carries a signed ProofOfAuditReport (kind 30105) binding the report hash and your bundle hash, and sha256(audit_hash|input_sha256|proof_id) is anchored in a 0-value Base transaction after the response (status at /v1/services/result/{job_id}/anchor, free). Pass `tier: diagnostic` for the R1+R2 subset, or buy the $0.25 ops.agent-workforce-diagnostic rung first. Bundle cap: 4 MB per request on this endpoint. Sending more than 4 MB of transcripts, or do not want to send them at all? Run the free local extractor https://craigmbrown.com/blindoracle/tools/workforce_extract.py (sha256 34f33389a564555e4d4476b86982feff10ca53eb6143bbc38139798d8dd09417) and send its `summary` instead of `records`: counts only, no message text, any number of days, identical numbers. The report then states that it was computed from a buyer-run extraction and whether your extractor matches the published hash; the attestation proves that input produced this output, not that the input is complete. (x402: $25.00 USDC per call)
Annotations the server declares
- title: Agent Workforce Early-Warning Report
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
ops.due-diligence-scanAutomated DD scan: financials, litigation, key personnel, IP, media sentiment, red flags Refuses, with no charge, when ≥5 web sources + registry/EDGAR lookups cannot be fetched.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Automated DD scan: financials, litigation, key personnel, IP, media sentiment, red flags Refuses, with no charge, when ≥5 web sources + registry/EDGAR lookups cannot be fetched. Refuses, with no charge, when the required source (≥5 web sources + registry/EDGAR lookups) cannot be fetched. (x402: $1.00 USDC per call)
Annotations the server declares
- title: Due Diligence Pre-Screening
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
ops.link-integrityChecks every URL you list in `prompt` (HEAD, then GET) and returns PASS/FAIL with each URL's status code.Reads text others writeNo findings
Why it is in this group
- The name "ops.link-integrity" says "link": a word for a change.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Checks every URL you list in `prompt` (HEAD, then GET) and returns PASS/FAIL with each URL's status code. It does not crawl a site: list every URL you want checked, for example after a deploy. (x402: $0.01 USDC per call)
Annotations the server declares
- title: Post-Deploy Link Integrity Check
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
oracle.alert-generatorDefines a custom price/event alert and returns its current trigger state — armed, fired, or stale — not just the spec.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Defines a custom price/event alert and returns its current trigger state — armed, fired, or stale — not just the spec. Settlement proof: ProofOfSettledOutcome (kind 30120, data/proof_settled_outcomes.jsonl). Reads off oracle.volatility-monitor and oracle.price-feed; this SKU is the trigger layer, not the data layer. Refuses, with no charge, when live price + buyer threshold fields cannot be fetched. (x402: $0.02 USDC per call)
Annotations the server declares
- title: Alert Generator
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
oracle.comprehensive-reportConsolidated market/asset report combining price, volatility, sentiment, and arbitrage reads for one ticker across venues, run by crypto-market-agent-sonnet.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Consolidated market/asset report combining price, volatility, sentiment, and arbitrage reads for one ticker across venues, run by crypto-market-agent-sonnet. Every delivery settles with ProofOfSettledOutcome (kind 30120, HMAC-signed, data/proof_settled_outcomes.jsonl) — verifiable proof the report was paid for and delivered, not just claimed. Broader scope than any single oracle.* SKU below; narrower and cheaper than security.massat-audit's compliance-grade evidence. May deliver with a first-line DEGRADED notice when live data is unavailable. (x402: $0.20 USDC per call)
Annotations the server declares
- title: Comprehensive Report
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
oracle.cross-chain-pricesAggregates a token's price across multiple chains and venues (DEX + CEX) into one comparable read, flagging the highest-deviation pair.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Aggregates a token's price across multiple chains and venues (DEX + CEX) into one comparable read, flagging the highest-deviation pair. Settlement proof: ProofOfSettledOutcome (kind 30120, data/proof_settled_outcomes.jsonl). Narrower than oracle.comprehensive-report (single metric, not a full report); feeds oracle.market-arbitrage's spread detection. Refuses, with no charge, when live price feed, ≥2 chains cannot be fetched. (x402: $0.03 USDC per call)
Annotations the server declares
- title: Cross-Chain Prices
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
oracle.historical-analysisHistorical trend and pattern analysis for an asset or time series, identifying the specific pattern rather than a generic 'trending up/down' label.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Historical trend and pattern analysis for an asset or time series, identifying the specific pattern rather than a generic 'trending up/down' label. Settlement proof: ProofOfSettledOutcome (kind 30120, data/proof_settled_outcomes.jsonl). Distinct from oracle.volatility-monitor's real-time (not historical) read. Refuses, with no charge, when price history ≥30d cannot be fetched. (x402: $0.08 USDC per call)
Annotations the server declares
- title: Historical Analysis
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
oracle.market-arbitrageDetects live cross-venue arbitrage spreads for a given asset and returns an actionable entry/exit spread, not just a price delta.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Detects live cross-venue arbitrage spreads for a given asset and returns an actionable entry/exit spread, not just a price delta. Built on oracle.cross-chain-prices' price aggregation. Settlement proof: ProofOfSettledOutcome (kind 30120, data/proof_settled_outcomes.jsonl). Refuses, with no charge, when live prices from ≥2 venues cannot be fetched. (x402: $0.10 USDC per call)
Annotations the server declares
- title: Market Arbitrage
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
oracle.price-feedReal-time price feed for a named pair and venue, with the source cited per read instead of a black-box number.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Real-time price feed for a named pair and venue, with the source cited per read instead of a black-box number. Settlement proof: ProofOfSettledOutcome (kind 30120, data/proof_settled_outcomes.jsonl). The base data layer the other oracle.* SKUs (arbitrage, volatility, alerts) are built on. Refuses, with no charge, when live price feed cannot be fetched. Refuses, with no charge, when the required source (live price feed) cannot be fetched. (x402: $0.02 USDC per call)
Annotations the server declares
- title: Oracle Price Feed
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
oracle.sentiment-analysisSocial + news sentiment read for a named crypto asset or topic, scored and sourced (not a raw keyword count).Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Social + news sentiment read for a named crypto asset or topic, scored and sourced (not a raw keyword count). Settlement proof: ProofOfSettledOutcome (kind 30120, data/proof_settled_outcomes.jsonl). Same sentiment methodology as research.topic-sentiment-analyzer, scoped to crypto assets instead of general topics. May deliver with a first-line DEGRADED notice when live data is unavailable. (x402: $0.06 USDC per call)
Annotations the server declares
- title: Sentiment Analysis
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
oracle.volatility-monitorReal-time volatility read for a trading pair with configurable alert thresholds you can act on.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Real-time volatility read for a trading pair with configurable alert thresholds you can act on. Settlement proof: ProofOfSettledOutcome (kind 30120, data/proof_settled_outcomes.jsonl). Complements oracle.alert-generator (fires the alert) and oracle.historical-analysis (explains the trend behind the number). Refuses, with no charge, when live price history cannot be fetched. Refuses, with no charge, when the required source (live price history) cannot be fetched. (x402: $0.04 USDC per call)
Annotations the server declares
- title: Volatility Monitor
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
procurement.council5-11 agent panel debate playing CFO + CIO + CISO + Procurement Lead.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it 5-11 agent panel debate playing CFO + CIO + CISO + Procurement Lead. Forced vote on any vendor decision. Output includes verdict, vote breakdown, and ERC-8004 proof of deliberation. Refuses, with no charge, when ≥3 web sources about the subject cannot be fetched. Refuses, with no charge, when the required source (≥3 web sources about the subject) cannot be fetched. (x402: $2.00 USDC per call)
Annotations the server declares
- title: Procurement Council on Demand
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
procurement.trust-layerSigned, ledger-derived trust evidence about a NAMED BlindOracle agent (pass the agent name or erc8004 id as `subject`): passport status, audit-report count (ProofOfAuditReport 3010…Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Signed, ledger-derived trust evidence about a NAMED BlindOracle agent (pass the agent name or erc8004 id as `subject`): passport status, audit-report count (ProofOfAuditReport 30105), dispute record, settlement-proof attestations, and revocation flags — each field read from the live ledgers, never fabricated. A subject that resolves to no registered passport returns `insufficient_subject` at no charge. It does NOT research external companies; use procurement.vendor-vetting or ops.due-diligence-scan for that. (x402: $0.01 USDC per call)
Annotations the server declares
- title: Procurement Trust Layer
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
procurement.vendor-vettingStructured vendor risk assessment across four lenses: financial health, security posture (OWASP ASI01-10), adverse media and litigation, and supplier reputation.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Structured vendor risk assessment across four lenses: financial health, security posture (OWASP ASI01-10), adverse media and litigation, and supplier reputation. Returns a 0-10 risk score, per-category findings, and red flags. Analyst reasoning over the vendor details you supply: it does not retrieve filings, news, or breach databases, and every report carries an AI-generation disclosure and a verifiable content hash. For an automated audit of a live agent system, see security.massat-audit. Refuses, with no charge, when ≥5 web sources incl. vendor's published terms cannot be fetched. (x402: $99.00 USDC per call)
Annotations the server declares
- title: AI Vendor Vetting
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
research.topic-deep-researcherStructured research brief (exec summary, mechanisms, alternatives, risks, [n] citations) synthesized over live web search results plus any `urls` you supply — each supplied URL is …Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Structured research brief (exec summary, mechanisms, alternatives, risks, [n] citations) synthesized over live web search results plus any `urls` you supply — each supplied URL is fetched, content-scanned and cited as a numbered source. It is a $0.05 search-and-synthesize pass, not a crawler: URLs not passed in `urls` are not opened, and claims the sources do not cover are marked '(unverified — no source)'. Settlement proof: ProofOfSettledOutcome (kind 30120). Deeper than research.topic-news-scanner; web + supplied documents vs content.youtube-research's video-only pool. Refuses, with no charge, when ≥5 fetched web sources cannot be fetched. (x402: $0.05 USDC per call)
Annotations the server declares
- title: Deep Topic Research
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
research.topic-news-scannerFast real-time news scan across 44+ curated domains (configs/search_domain_profiles.json), every item source-attributed.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Fast real-time news scan across 44+ curated domains (configs/search_domain_profiles.json), every item source-attributed. Settlement proof: ProofOfSettledOutcome (kind 30120, data/proof_settled_outcomes.jsonl). The quick-scan tier below research.topic-deep-researcher's slower, citation-grade report. Refuses, with no charge, when ≥3 fetched web sources cannot be fetched. Refuses, with no charge, when the required source (≥3 fetched web sources) cannot be fetched. (x402: $0.02 USDC per call)
Annotations the server declares
- title: News Intelligence Scanner
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
research.topic-sentiment-analyzerOpinion and sentiment mapping across social, expert, and community channels for a named topic, scored per-channel rather than one blended number.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Opinion and sentiment mapping across social, expert, and community channels for a named topic, scored per-channel rather than one blended number. Settlement proof: ProofOfSettledOutcome (kind 30120, data/proof_settled_outcomes.jsonl). Same methodology as oracle.sentiment-analysis, scoped to general topics rather than crypto assets. Refuses, with no charge, when ≥3 fetched web sources cannot be fetched. Refuses, with no charge, when the required source (≥3 fetched web sources) cannot be fetched. (x402: $0.03 USDC per call)
Annotations the server declares
- title: Sentiment Analysis
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
security.audit-attestationNotarization of an AI audit result by BlindOracle (not a neutral third party: we operate this marketplace): we did not run the audit, we attest that a specified audit was performed…Reads text others write1 finding
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Findings
- Info
The description mentions "credential".
What this check looks forMatched text
description: "...its findings match what's claimed — the credential a checkpoint can trust without re-runni..."
Description, as the server wrote it Notarization of an AI audit result by BlindOracle (not a neutral third party: we operate this marketplace): we did not run the audit, we attest that a specified audit was performed and its findings match what's claimed — the credential a checkpoint can trust without re-running the audit itself. (x402: $0.50 USDC per call)
Annotations the server declares
- title: AI Audit Attestation (Notary)
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
security.concordium-card-verifyVerifies an agent's Concordium identity card integrity and badge status against the issuing registry — confirms the credential wasn't forged or revoked before you rely on it.Reads text others write1 finding
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Findings
- Info
The description mentions "credential".
What this check looks forMatched text
description: "...nst the issuing registry — confirms the credential wasn't forged or revoked before you rel..."
Description, as the server wrote it Verifies an agent's Concordium identity card integrity and badge status against the issuing registry — confirms the credential wasn't forged or revoked before you rely on it. (x402: $0.02 USDC per call)
Annotations the server declares
- title: Concordium Agent Card Integrity + Badge Check
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
security.enterprise-audit13-agent coordinated security audit producing a signed ProofOfAuditReport, Merkle-anchored to Base, for enterprises that need documented evidence an agent fleet was assessed before…Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it 13-agent coordinated security audit producing a signed ProofOfAuditReport, Merkle-anchored to Base, for enterprises that need documented evidence an agent fleet was assessed before deployment — not a self-attestation. Refuses, with no charge, when the required source (buyer target) cannot be fetched. (x402: $25.00 USDC per call)
Annotations the server declares
- title: Enterprise AI Security Audit (13-agent)
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
security.injection-resilienceScans text you supply (an agent's system prompt, tool descriptions, manifest, or output it produced) for prompt-injection and content-trap patterns (OWASP ASI01), using the same de…Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Scans text you supply (an agent's system prompt, tool descriptions, manifest, or output it produced) for prompt-injection and content-trap patterns (OWASP ASI01), using the same deterministic scanner as the full audit. It does not send probes to the agent; for an attested ASI01-10 audit use security.massat-audit. (x402: $0.50 USDC per call)
Annotations the server declares
- title: Prompt-Injection Resilience Check
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
security.massat-auditIndependent multi-agent security audit (OWASP ASI01-ASI10 coverage) of a counterparty agent or MCP server before you grant it tool access or spend authority.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Independent multi-agent security audit (OWASP ASI01-ASI10 coverage) of a counterparty agent or MCP server before you grant it tool access or spend authority. Findings are Merkle-committed and the audit report is signed. (x402: $5.00 USDC per call)
Annotations the server declares
- title: Multi-Agent Security Audit
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
security.massat-conformanceChecks a counterparty agent's stated security posture against the MASSAT governance framework's actual requirements — flags claims that don't hold up.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Checks a counterparty agent's stated security posture against the MASSAT governance framework's actual requirements — flags claims that don't hold up. (x402: $0.25 USDC per call)
Annotations the server declares
- title: MASSAT Governance Conformance Check
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
security.process-attestationSigned attestation that a specific process was followed (not just that an outcome occurred) — useful when a counterparty needs evidence of *how* something was done, not only *that*…Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Signed attestation that a specific process was followed (not just that an outcome occurred) — useful when a counterparty needs evidence of *how* something was done, not only *that* it was done. (x402: $0.25 USDC per call)
Annotations the server declares
- title: Process-Followed Attestation
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
translation.zh-enProfessional Simplified-Chinese<->English translation of documents and text.Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Professional Simplified-Chinese<->English translation of documents and text. Auto-detects direction, preserves formatting, names, and numbers. Offshore/PIPL-guarded (refuses mainland-CN data subjects). BO trust envelope (content hash + scan + provenance) on every deliverable. (x402: $0.05 USDC per call)
Annotations the server declares
- title: Chinese<->English Translation
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
trust.receipt-verifyVerify a BlindOracle receipt against the parts a buyer cannot check alone: is the proof_id really in BlindOracle's append-only proof rail, does its payload still hash to the stored…Reads text others writeNo findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Verify a BlindOracle receipt against the parts a buyer cannot check alone: is the proof_id really in BlindOracle's append-only proof rail, does its payload still hash to the stored row, and does the row's HMAC-SHA256 verify against the signing key. Also recomputes content_sha256 and confirms the Base payment, so one call is a complete answer. Deterministic, no LLM. The self-checkable half is free forever in the published verify-blindoracle-receipt skill. Called with a delegation_token (the delegator's) and the deliverable_sha256 you received, it closes a delegation chain instead: a signed AgentDelegationClosure saying whether what you received matches what the delegate sealed. (x402: $0.01 USDC per call)
Annotations the server declares
- title: BlindOracle Receipt Verification (seller-side)
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
Only reads 2Show 2 read-only tools
list_servicesFREE, no wallet, no API key, arguments optional: every BlindOracle service with its price, which tools are free, how many free calls you have left today, and how to pay (x402 USDC …Reads private dataNo findings
Why it is in this group
- The name "list_services" says "list": a word for reading.
What it can reach
- private-dataThe description's first sentence says "note": it reads notes.
Description, as the server wrote it FREE, no wallet, no API key, arguments optional: every BlindOracle service with its price, which tools are free, how many free calls you have left today, and how to pay (x402 USDC on Base, or a free starter-credit note). Call this first to see what is available.
Annotations the server declares
- title: List services and prices
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
reputation.lookupFREE over MCP: 5 calls per day in total, one pool shared by agent_trust-badge and reputation_lookup; no wallet, no API key.Reads text others writeNo findings
Why it is in this group
- The name "reputation.lookup" says "lookup": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it FREE over MCP: 5 calls per day in total, one pool shared by agent_trust-badge and reputation_lookup; no wallet, no API key. A lookup that finds nothing does not use one. Free only when the call carries NO payment: a call that sends a payment or starter-credit note is charged. Look up an agent's BlindOracle reputation before you transact with it: score, level and badge, plus counts of its recorded runs (successful and failed). Run counts are the agent's own executions, not settled marketplace jobs; an agent with no history scores zero. After the free calls: $0.01 USDC via x402 (network eip155:8453).
Annotations the server declares
- title: Agent Reputation Lookup
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: false
- openWorldHint: true
This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.
Exfil surface
Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.
No private-data, untrusted-input, outbound path in this server alone: no tool is tagged outbound.
- Reads private data1
- Reads files, mail, messages, databases, secrets, calendars or repositories.
- Reads text others write53
- Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
- Can send data out0
- Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.
Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.
Remote posture
What anyone on the internet sees at the remote endpoint: an anonymous initialize and tools/list, and the public OAuth metadata.
Looked at Nov 20, 2026, 19:03 UTC, unchanged since Nov 6, 2026. No credentials were sent and no tool was called. What is looked at.
- Reachable
- yes
- HTTP status of an anonymous initialize
- 200
- tools/list served without credentials
- yes
- Tools that change things, listed without credentials
- 53
- Protocol version
- 2025-06-18
- WWW-Authenticate on 401
- not observed
- resource_metadata in WWW-Authenticate
- not observed
- RFC 9728 protected resource metadata
- https://api.craigmbrown.com/.well-known/oauth-protected-resource
- Authorization servers
- https://api.craigmbrown.com
- Legacy HTTP+SSE transport
- no
- Dynamic client registration
- yes
- Client ID metadata documents
- no
- WarnThe whole server
The endpoint served tools/list without credentials, and the list has 53 tools that change things (53 write). Whether a call also works without credentials was not tested: tools are never called.
What this check looks forMatched text
https://api.craigmbrown.com/v2/?src=mcp_official_registry: tools/list returned 55 tools to an anonymous client, including health_check (write), agent.prehire-check (write), agent.trust-badge (write), arbitration.dispute-settlement (write), attestation.single-use-seal (write) and 48 more
Posture changes
- Nov 6, 2026
- Tools that change things, listed without credentials: 40 to 53
- Oct 29, 2026
- Tools that change things, listed without credentials: 38 to 40
- RFC 9728 protected resource metadata: none to https://api.craigmbrown.com/.well-known/oauth-protected-resource
- Authorization servers: none to https://api.craigmbrown.com
- Dynamic client registration: not observed to yes
- Client ID metadata documents: not observed to no
Findings
Matches of fixed checks, with the text they matched.
The description mentions "credential".
What this check looks forMatched text
description: "...gated a task to a named agent: a signed credential (did:web:craigmbrown.com#delegation-1,..."
The description is 2,079 characters long, over the 2,000 this check looks for.
What this check looks forMatched text
description ends: "...roduced this output, not that the input is complete. (x402: $25.00 USDC per call)"
The description mentions "credential".
What this check looks forMatched text
description: "...its findings match what's claimed — the credential a checkpoint can trust without re-runni..."
The description mentions "credential".
What this check looks forMatched text
description: "...nst the issuing registry — confirms the credential wasn't forged or revoked before you rel..."
History
Changes to its tool definitions and to its remote endpoint, newest first.
Tools2 added, 1 removed, 53 changed.+2 added−1 removed~53 changedView diff
- Added:bundle.smb-agent-guardrailsChanges things
- Added:list_servicesOnly reads
- Removed:prediction.blindoracle
- Changed:agent.prehire-checkChanges thingsannotations
- Changed:agent.trust-badgeChanges thingsdescriptionannotations
- Changed:arbitration.dispute-settlementChanges thingsannotations
- Changed:attestation.single-use-sealChanges thingsannotations
- Changed:bundle.delegation-chainChanges thingsannotations
- Changed:bundle.dispute-evidenceChanges thingsannotations
- Changed:bundle.due-diligenceChanges thingsannotations
- Changed:bundle.enterprise-ladderChanges thingsannotations
- Changed:bundle.massat-ladderChanges thingsannotations
- Changed:bundle.prehire-ladderChanges thingsannotations
- Changed:bundle.reputation-badgeChanges thingsannotations
- Changed:bundle.researched-debateChanges thingsannotations
- Changed:bundle.seal-and-badgeChanges thingsannotations
- Changed:bundle.trust-paperworkChanges thingsannotations
- Changed:bundle.vendor-vettingChanges thingsdescriptionannotations
- Changed:content.youtube-researchChanges thingsannotations
- Changed:crypto.market-analyzerChanges thingsdescriptionannotations
- Changed:data.business-registryChanges thingsannotations
- Changed:data.sec-edgar-filingChanges thingsannotations
- Changed:data.web-extractChanges thingsannotations
- Changed:delegation.acceptChanges thingsannotations
- Changed:delegation.recordChanges thingsdescriptionannotations
- Changed:deliberation.multi-agent-debateChanges thingsannotations
- Changed:finops.token-spend-auditChanges thingsannotations
- Changed:health_checkChanges thingsannotations
- Changed:ops.agent-workforce-diagnosticChanges thingsannotations
- Changed:ops.agent-workforce-reportChanges thingsannotations
- Changed:ops.due-diligence-scanChanges thingsannotations
- Changed:ops.link-integrityChanges thingsdescriptionannotations
- Changed:oracle.alert-generatorChanges thingsannotations
- Changed:oracle.comprehensive-reportChanges thingsannotations
- Changed:oracle.cross-chain-pricesChanges thingsannotations
- Changed:oracle.historical-analysisChanges thingsdescriptionannotations
- Changed:oracle.market-arbitrageChanges thingsdescriptionannotations
- Changed:oracle.price-feedChanges thingsannotations
- Changed:oracle.sentiment-analysisChanges thingsannotations
- Changed:oracle.volatility-monitorChanges thingsannotations
- Changed:procurement.councilChanges thingsannotations
- Changed:procurement.trust-layerChanges thingsannotations
- Changed:procurement.vendor-vettingChanges thingsannotations
- Changed:reputation.lookupOnly readsdescriptionannotations
- Changed:research.topic-deep-researcherChanges thingsannotations
- Changed:research.topic-news-scannerChanges thingsannotations
- Changed:research.topic-sentiment-analyzerChanges thingsannotations
- Changed:security.audit-attestationChanges thingsdescriptionannotations
- Changed:security.concordium-card-verifyChanges thingsannotations
- Changed:security.enterprise-auditChanges thingsannotations
- Changed:security.injection-resilienceChanges thingsannotations
- Changed:security.massat-auditChanges thingsannotations
- Changed:security.massat-conformanceChanges thingsannotations
- Changed:security.process-attestationChanges thingsannotations
- Changed:translation.zh-enChanges thingsannotations
- Changed:trust.receipt-verifyChanges thingsdescriptionannotations
PostureRemote posture changed. Tools that change things, listed without credentials: 40 -> 53.View diff
- Tools that change things, listed without credentials: 40 to 53
Tools3 added, 2 changed.+3 added~2 changedView diff
- Added:bundle.delegation-chainChanges things
- Added:delegation.acceptChanges things
- Added:delegation.recordChanges things
- Changed:attestation.single-use-sealChanges thingsdescription
- Changed:trust.receipt-verifyChanges thingsdescription
Tools12 added, 3 changed.+12 added~3 changedView diff
- Added:bundle.dispute-evidenceChanges things
- Added:bundle.due-diligenceChanges things
- Added:bundle.enterprise-ladderChanges things
- Added:bundle.massat-ladderChanges things
- Added:bundle.prehire-ladderChanges things
- Added:bundle.reputation-badgeChanges things
- Added:bundle.researched-debateChanges things
- Added:bundle.seal-and-badgeChanges things
- Added:bundle.trust-paperworkChanges things
- Added:bundle.vendor-vettingChanges things
- Added:ops.agent-workforce-diagnosticChanges things
- Added:ops.agent-workforce-reportChanges things
- Changed:agent.trust-badgeChanges thingsdescription
- Changed:reputation.lookupOnly readsdescription
- Changed:security.injection-resilienceChanges thingsdescription
PostureRemote posture changed. Tools that change things, listed without credentials: 38 -> 40; RFC 9728 protected resource metadata: none -> https://api.craigmbrown.com/.well-known/oauth-protected-resource; Authorization servers: none -> https://a…View diff
- Tools that change things, listed without credentials: 38 to 40
- RFC 9728 protected resource metadata: none to https://api.craigmbrown.com/.well-known/oauth-protected-resource
- Authorization servers: none to https://api.craigmbrown.com
- Dynamic client registration: not observed to yes
- Client ID metadata documents: not observed to no
Install
Pinned to version 2.0.5. Values in angle brackets are yours to fill in.
claude mcp add --transport http blindoracle https://api.craigmbrown.com/v2/?src=mcp_official_registryclaude mcp add --transport http blindoracle https://api.craigmbrown.com/v2/?src=mcp_official_registryRun in a terminal.
Claude Code permissions
Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.
{
"permissions": {
"allow": [
"mcp__blindoracle__list_services",
"mcp__blindoracle__reputation_lookup"
],
"ask": [
"mcp__blindoracle__health_check",
"mcp__blindoracle__agent_prehire-check",
"mcp__blindoracle__agent_trust-badge",
"mcp__blindoracle__arbitration_dispute-settlement",
"mcp__blindoracle__attestation_single-use-seal",
"mcp__blindoracle__bundle_delegation-chain",
"mcp__blindoracle__bundle_dispute-evidence",
"mcp__blindoracle__bundle_due-diligence",
"mcp__blindoracle__bundle_enterprise-ladder",
"mcp__blindoracle__bundle_massat-ladder",
"mcp__blindoracle__bundle_prehire-ladder",
"mcp__blindoracle__bundle_reputation-badge",
"mcp__blindoracle__bundle_researched-debate",
"mcp__blindoracle__bundle_seal-and-badge",
"mcp__blindoracle__bundle_smb-agent-guardrails",
"mcp__blindoracle__bundle_trust-paperwork",
"mcp__blindoracle__bundle_vendor-vetting",
"mcp__blindoracle__content_youtube-research",
"mcp__blindoracle__crypto_market-analyzer",
"mcp__blindoracle__data_business-registry",
"mcp__blindoracle__data_sec-edgar-filing",
"mcp__blindoracle__data_web-extract",
"mcp__blindoracle__delegation_accept",
"mcp__blindoracle__delegation_record",
"mcp__blindoracle__deliberation_multi-agent-debate",
"mcp__blindoracle__finops_token-spend-audit",
"mcp__blindoracle__ops_agent-workforce-diagnostic",
"mcp__blindoracle__ops_agent-workforce-report",
"mcp__blindoracle__ops_due-diligence-scan",
"mcp__blindoracle__ops_link-integrity",
"mcp__blindoracle__oracle_alert-generator",
"mcp__blindoracle__oracle_comprehensive-report",
"mcp__blindoracle__oracle_cross-chain-prices",
"mcp__blindoracle__oracle_historical-analysis",
"mcp__blindoracle__oracle_market-arbitrage",
"mcp__blindoracle__oracle_price-feed",
"mcp__blindoracle__oracle_sentiment-analysis",
"mcp__blindoracle__oracle_volatility-monitor",
"mcp__blindoracle__procurement_council",
"mcp__blindoracle__procurement_trust-layer",
"mcp__blindoracle__procurement_vendor-vetting",
"mcp__blindoracle__research_topic-deep-researcher",
"mcp__blindoracle__research_topic-news-scanner",
"mcp__blindoracle__research_topic-sentiment-analyzer",
"mcp__blindoracle__security_audit-attestation",
"mcp__blindoracle__security_concordium-card-verify",
"mcp__blindoracle__security_enterprise-audit",
"mcp__blindoracle__security_injection-resilience",
"mcp__blindoracle__security_massat-audit",
"mcp__blindoracle__security_massat-conformance",
"mcp__blindoracle__security_process-attestation",
"mcp__blindoracle__translation_zh-en",
"mcp__blindoracle__trust_receipt-verify"
],
"deny": []
}
}Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as blindoracle. Review it before you use it.
Capability tags behind 54 rules
- mcp__blindoracle__list_services private-data
- mcp__blindoracle__agent_prehire-check untrusted-input
- mcp__blindoracle__agent_trust-badge untrusted-input
- mcp__blindoracle__arbitration_dispute-settlement untrusted-input
- mcp__blindoracle__attestation_single-use-seal untrusted-input
- mcp__blindoracle__bundle_delegation-chain untrusted-input
- mcp__blindoracle__bundle_dispute-evidence untrusted-input
- mcp__blindoracle__bundle_due-diligence untrusted-input
- mcp__blindoracle__bundle_enterprise-ladder untrusted-input
- mcp__blindoracle__bundle_massat-ladder untrusted-input
- mcp__blindoracle__bundle_prehire-ladder untrusted-input
- mcp__blindoracle__bundle_reputation-badge untrusted-input
- mcp__blindoracle__bundle_researched-debate untrusted-input
- mcp__blindoracle__bundle_seal-and-badge untrusted-input
- mcp__blindoracle__bundle_smb-agent-guardrails untrusted-input
- mcp__blindoracle__bundle_trust-paperwork untrusted-input
- mcp__blindoracle__bundle_vendor-vetting untrusted-input
- mcp__blindoracle__content_youtube-research untrusted-input
- mcp__blindoracle__crypto_market-analyzer untrusted-input
- mcp__blindoracle__data_business-registry untrusted-input
- mcp__blindoracle__data_sec-edgar-filing untrusted-input
- mcp__blindoracle__data_web-extract untrusted-input
- mcp__blindoracle__delegation_accept untrusted-input
- mcp__blindoracle__delegation_record untrusted-input
- mcp__blindoracle__deliberation_multi-agent-debate untrusted-input
- mcp__blindoracle__finops_token-spend-audit untrusted-input
- mcp__blindoracle__ops_agent-workforce-diagnostic untrusted-input
- mcp__blindoracle__ops_agent-workforce-report untrusted-input
- mcp__blindoracle__ops_due-diligence-scan untrusted-input
- mcp__blindoracle__ops_link-integrity untrusted-input
- mcp__blindoracle__oracle_alert-generator untrusted-input
- mcp__blindoracle__oracle_comprehensive-report untrusted-input
- mcp__blindoracle__oracle_cross-chain-prices untrusted-input
- mcp__blindoracle__oracle_historical-analysis untrusted-input
- mcp__blindoracle__oracle_market-arbitrage untrusted-input
- mcp__blindoracle__oracle_price-feed untrusted-input
- mcp__blindoracle__oracle_sentiment-analysis untrusted-input
- mcp__blindoracle__oracle_volatility-monitor untrusted-input
- mcp__blindoracle__procurement_council untrusted-input
- mcp__blindoracle__procurement_trust-layer untrusted-input
- mcp__blindoracle__procurement_vendor-vetting untrusted-input
- mcp__blindoracle__reputation_lookup untrusted-input
- mcp__blindoracle__research_topic-deep-researcher untrusted-input
- mcp__blindoracle__research_topic-news-scanner untrusted-input
- mcp__blindoracle__research_topic-sentiment-analyzer untrusted-input
- mcp__blindoracle__security_audit-attestation untrusted-input
- mcp__blindoracle__security_concordium-card-verify untrusted-input
- mcp__blindoracle__security_enterprise-audit untrusted-input
- mcp__blindoracle__security_injection-resilience untrusted-input
- mcp__blindoracle__security_massat-audit untrusted-input
- mcp__blindoracle__security_massat-conformance untrusted-input
- mcp__blindoracle__security_process-attestation untrusted-input
- mcp__blindoracle__translation_zh-en untrusted-input
- mcp__blindoracle__trust_receipt-verify untrusted-input
Pin
Keep the decision: fail CI when the tools change, follow every change, or show a badge.
Pin in CI
Fail a build when the tools you reviewed are not the tools the server serves today.
curl -s "https://radar.anlyon.com/api/v1/servers/io.github.craigmbrown/blindoracle/verify?hash=2bcb255207d58909b23bb6302c74f05d30449905192b8f356df8e845c15836a9" | jq -e '.changed == false'Pinned hash 2bcb255207d5, from Nov 19, 2026, 17:18 UTC.
Watch
Get told when its tools change: a new tool, a new description, a new class.
https://radar.anlyon.com/servers/io.github.craigmbrown/blindoracle/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://radar.anlyon.com/servers/io.github.craigmbrown/blindoracle)Source: MCP Registry · repository. Synced Nov 5, 2026, 19:17 UTC.
blindoracle MCP server: questions and answers
Answered from the scan above, not from the publisher's description.
Is the blindoracle MCP server safe?
- No scan can promise that. What its definitions show: 55 tools, 53 of which can change, send or delete something, and 4 findings from fixed checks. Allow the reads, ask before each change, and deny the deletes; the permissions block on this page does that.
Can the blindoracle MCP server delete data?
- No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.
Can the blindoracle MCP server do things that cannot be undone?
- No tool is classed as irreversible: none of them sends, pays, publishes or merges, going by their definitions.
What tools does the blindoracle MCP server have?
- 55 tools: 53 write, 2 read. health_check, agent.prehire-check, agent.trust-badge, arbitration.dispute-settlement, attestation.single-use-seal, bundle.delegation-chain, bundle.dispute-evidence, bundle.due-diligence and 47 more.
Does the blindoracle MCP server need sign-in or OAuth?
- Its remote endpoint listed its tools to an anonymous client, without sign-in.
Does the blindoracle MCP server send data to outside services?
- None of its tools sends, posts or publishes, or takes a recipient, channel or URL.
Has the blindoracle MCP server changed its tools?
- Yes, 3 times since Anlyon Radar started tracking it, most recently on Nov 19, 2026. Every change is on this page and in its Atom feed.