Skip to content

PG1 Sovereign Threat Intelligence MCP server

io.github.Project-Gifted1/pg1-threat-intelProject-Gifted1streamable-httpv1.22.0Scanned

Preflight for AI agents, signed: one call; wallet, domain, IP abuse, package checks and CVEs.

22 tools: 14 change things and 8 only read.

Tools: 0 delete, 0 irreversible, 14 write, 8 read.
  • Reads private data · 0
  • Reads text others write · 1
  • Can send data out · 3
  • 8 findings, 3 high

Sign-in required, no OAuth metadata. Remote posture

With Anlyon

14 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.
Govern this server →

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 PG1 Sovereign Threat Intelligence MCP server can do

The PG1 Sovereign Threat Intelligence MCP server is published by the GitHub account Project-Gifted1. It has 22 tools. It can create or change data (14 tools: preflight_check, check_x402_endpoint, check_agent_card, check_agent_message and 10 more), but cannot delete data. 8 tools only read.

1 tool brings in text other people can write (get_ioc_context); 3 tools can send data out of your system (check_x402_endpoint, check_agent_card and check_agent_message).

The fixed checks matched 8 findings, 3 of them high severity (model-directed-text, sensitive-path, long-description and credentials-mention). Each one comes with the text it matched.

Its tools changed 2 times since Anlyon Radar first saw them.

Last scanned Nov 21, 2026, version 1.22.0.

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.

22 tools

Changes things 14Claude Code: ask
  • check_agent_cardPG1 Sovereign Threat Intelligence: checks another agent before you trust it.Can send data outNo 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

    • outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.
    Description, as the server wrote it
    PG1 Sovereign Threat Intelligence: checks another agent before you trust it. Fetches its /.well-known/agent-card.json (https only; private addresses and redirects refused): is it a valid A2A agent card, does its name claim a well-known brand the domain does not belong to, how old is the domain, is the hostname on the phishing blocklist or a look-alike. Always free (60/hour per caller). Give url or domain. Returns status flagged / no_flags / unknown; no_flags means no flags found, never "safe".
  • check_agent_messagePG1 Sovereign Threat Intelligence: a pre-read check for a message from another agent or a seller's skill.md.Can send data out3 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

    • outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.

    Findings

    • High

      The description contains "ignore previous instructions", an instruction to set earlier instructions aside.

      Matched text

      description: "...: instruction override and role hijack (ignore previous instructions, "you are now", developer or jailbreak..."

      What this check looks for
    • High

      The description refers to ".env" and "~/.ssh".

      Matched text

      description: "...ironment variables, keys, seed phrases, .env or ~/.ssh; image or link URLs carrying..."

      What this check looks for
    • Info

      The description is 2,331 characters long, over the 2,000 this check looks for.

      Matched text

      description ends: "...gth, kind, verdict and reason codes are logged. Always free (60/hour per caller)."

      What this check looks for
    Description, as the server wrote it
    PG1 Sovereign Threat Intelligence: a pre-read check for a message from another agent or a seller's skill.md. Call it BEFORE your own model reads the content, to catch prompt injection and wallet or payment manipulation. Give exactly one of text (up to 64 KB) or url (https only; fetched once with the same private-address and no-redirect rules as check_x402_endpoint, 6 s timeout, text/markdown or text/plain only, up to 64 KB), and optionally kind (skill_md, message or other). PG1's own deterministic rules, no AI model: instruction override and role hijack (ignore previous instructions, "you are now", developer or jailbreak modes, claims to come from a system, an admin, an AI company or the wallet's developer); hidden content (zero-width and bidirectional control characters, Unicode tag characters, instructions in HTML comments, mixed-script look-alike words, long base64 or hex blobs, data: URIs; base64, hex and tag-character text is decoded and checked too); exfiltration (requests to reveal or send environment variables, keys, seed phrases, .env or ~/.ssh; image or link URLs carrying templated or secret-like values; curl | sh); wallet and payment manipulation (sign, approve, transfer or bridge instructions, unlimited allowances, a changed payTo or recipient, raised spending limits, presign-guard or other checks turned off); tool hijack (use your own tools or shell, install packages). Then, within 4 s, up to 5 domains and 5 wallet addresses named in the content go through check_domain_age, check_hostname_reputation, check_wallet_sanctions and check_wallet_age; if that does not finish, the rule results come back anyway (AGENT_MESSAGE_ENRICHMENT_INCOMPLETE). Returns verdict malicious (any high-severity finding: hidden tag characters, a secret-exfiltration instruction, a payment redirect, a sanctioned address), suspicious (medium findings only) or no_findings ("checked: no findings"), each finding with its reason code, severity, line and a masked excerpt (120 characters at most; keys, tokens and seed phrases masked), plus the URLs, domains and wallet addresses found and the sha256 and byte length of what was checked. No findings does not prove the content is harmless. The content is never stored; only its sha256, length, kind, verdict and reason codes are logged. Always free (60/hour per caller).
  • check_domain_agePG1 Sovereign Threat Intelligence: looks up a domain's registration age via RDAP (the IANA-standardized WHOIS successor), resolved through the IANA bootstrap registry for the corre…No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    Description, as the server wrote it
    PG1 Sovereign Threat Intelligence: looks up a domain's registration age via RDAP (the IANA-standardized WHOIS successor), resolved through the IANA bootstrap registry for the correct per-TLD RDAP server. No payment required — this tool is always free. SIBLING DIFFERENTIATION: Use for domain registration/age checks only. Do NOT use for reputation/threat-feed lookups (use get_ioc_context) or sanctions screening (use check_wallet_sanctions). BEHAVIOR: Returns { found: true, available: true, registration_date, age_days, expiration_date, registrar, newly_registered, source } when available, or { found: false, available: false, reason, reason_code } when the lookup does not resolve — this tool never estimates or guesses an age. "available" is a deprecated alias of "found", kept for backward compatibility. reason_code is "unsupported_tld" when the TLD has no RDAP server in the IANA bootstrap registry, or "timeout" / "lookup_failed" for other lookup failures. A newly registered domain (age_days < 30) is reported as a common phishing signal, not as proof of malicious intent.
  • check_hostname_reputationPG1 Sovereign Threat Intelligence: checks a single hostname against the MetaMask eth-phishing-detect blocklist/allowlist and a lookalike/typosquat detector, synced daily by the sov…No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    Description, as the server wrote it
    PG1 Sovereign Threat Intelligence: checks a single hostname against the MetaMask eth-phishing-detect blocklist/allowlist and a lookalike/typosquat detector, synced daily by the sovereign-threat-pipeline. No payment required — this tool is always free. SIBLING DIFFERENTIATION: Use for phishing/lookalike-domain screening of a hostname only. Do NOT use for domain registration age (use check_domain_age), general threat-feed indicator lookups (use get_ioc_context), or wallet sanctions screening (use check_wallet_sanctions). BEHAVIOR: Returns { hostname, verdict, sources, lookalike_of, list_synced_at, checked_at, attribution }. verdict is one of "allowlisted", "listed", "lookalike", or "not_listed" — this tool never returns "safe" or "clean", and a not_listed result means the hostname is not on the eth-phishing-detect lists, not that it is safe. "listed" results include match_type "exact" or "parent_domain" in sources. "lookalike" flags a probable typosquat/homoglyph of a known brand — via confusable-character skeleton matching within the stored tolerance, or a brand keyword embedded with extra words (e.g. metamask-login.com) — even when the hostname itself is not directly listed, and sets lookalike_of to the matched brand domain. A brand's own real domain or a subdomain of it is never flagged as its own lookalike. VALIDATION: accepts exactly one bare hostname per call (no bulk input); a value containing a URL scheme, path, port, spaces, or a wildcard returns an MCP tool error (isError: true, code invalid_hostname) instead of a verdict. Fails loudly (returns an error) if the phishing list data is unreachable or times out, rather than ever reporting not_listed on a data failure. Rate-limited to 60 calls/hour per caller when unauthenticated; a valid Gumroad license key (X-API-KEY header) exempts the limit, same as check_domain_age. List contents are never exposed beyond the single matched entry.
  • check_ip_abusePG1 Sovereign Threat Intelligence: looks up one public IPv4 or IPv6 address in AbuseIPDB using YOUR OWN AbuseIPDB API key (free or paid), and returns its abuse confidence score and…1 finding

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.

    Findings

    • Info

      The description is 2,321 characters long, over the 2,000 this check looks for.

      Matched text

      description ends: "...nauthenticated; a valid Gumroad license key (X-API-KEY header) exempts the limit."

      What this check looks for
    Description, as the server wrote it
    PG1 Sovereign Threat Intelligence: looks up one public IPv4 or IPv6 address in AbuseIPDB using YOUR OWN AbuseIPDB API key (free or paid), and returns its abuse confidence score and report counts with attribution to AbuseIPDB. BRING YOUR OWN KEY: send the key in the X-AbuseIPDB-Key HTTP request header on /api/mcp or /api/a2a - never as a tool argument. PG1 has no AbuseIPDB key of its own; without the header the call returns an MCP tool error (isError: true, code abuseipdb_key_required) and AbuseIPDB is not contacted. No PG1 payment required - lookups count against your own AbuseIPDB quota. SIBLING DIFFERENTIATION: Use for a live AbuseIPDB report lookup on a single IP only. Do NOT use for PG1's own threat-feed lookups (use get_ioc_context), hostnames (use check_hostname_reputation) or domains (use check_domain_age). BEHAVIOR: Returns { ip, abuse_confidence_score, total_reports, distinct_reporters, last_reported_at, country_code, usage_type, isp, domain, is_tor, is_whitelisted, attribution }. STATUS: "no_flags" when AbuseIPDB marks the address whitelisted (informational reason IP_WHITELISTED: AbuseIPDB does not count its reports against it); otherwise "flagged" when abuse_confidence_score > 0 (reason IP_ABUSE_REPORTED, with the score, report and reporter counts); otherwise "no_flags" (with informational reason IP_REPORTS_SCORED_ZERO when reports exist but AbuseIPDB scores them 0); "unknown" when the score or report count is missing. IP_TOR_EXIT_NODE is informational. abuse_confidence_score is AbuseIPDB's 0-100 confidence that the address is abusive; a score of 0 is not proof the address is harmless - this tool never returns "safe" or "clean". Results are returned only to the caller whose key was used, are never stored or shared, and are cached in memory for 15 minutes per key. ERRORS (isError: true): abuseipdb_key_required, invalid_abuseipdb_key (AbuseIPDB rejected the key), abuseipdb_rate_limited (your AbuseIPDB quota; retry_after passed through when AbuseIPDB sends it), upstream_unavailable (AbuseIPDB error or timeout), invalid_ip (not exactly one public address: private, reserved or malformed input is rejected without calling AbuseIPDB), invalid_max_age. Rate-limited to 60 calls/hour per caller when unauthenticated; a valid Gumroad license key (X-API-KEY header) exempts the limit.
  • check_packagePG1 Sovereign Threat Intelligence: a pre-install check for one npm or PyPI package - check before you install.1 finding

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.

    Findings

    • Info

      The description is 2,505 characters long, over the 2,000 this check looks for.

      Matched text

      description ends: "...mit. Data sources and licences: https://pg1-ai-agent.vercel.app/docs/attributions"

      What this check looks for
    Description, as the server wrote it
    PG1 Sovereign Threat Intelligence: a pre-install check for one npm or PyPI package - check before you install. No payment required - this tool is always free, and no AI model is involved. SIBLING DIFFERENTIATION: Use before installing or recommending a software package (npm or PyPI) only. Do NOT use for domains (use check_domain_age / check_hostname_reputation), CVE ids (use get_cve_details) or products by vendor name (use get_cve_by_product). BEHAVIOR: Returns { ecosystem, name, version_checked, exists, version_exists, first_published, age_days, latest_version, latest_published, reported_malicious, malicious_reports, vulnerabilities, lookalike_of, deprecated, install_scripts, attribution }. Signals: whether the package (and the version, if given) exists in the registry - a package that does not exist is a key signal, since names AI assistants invent get registered by attackers; first and latest publish dates; public malicious-package reports (OpenSSF Malicious Packages, via OSV.dev) for the version checked; known vulnerabilities for that version (OSV.dev: ids and severity only); whether the name looks like a popular package's (typosquat patterns); and, for npm only, the deprecated flag and install scripts (preinstall/install/postinstall). With no version, the latest version is checked. STATUS: "flagged" when the version is reported malicious (PACKAGE_REPORTED_MALICIOUS - worded "reported malicious" because such reports can be false positives), the package or version does not exist (PACKAGE_NOT_FOUND / PACKAGE_VERSION_NOT_FOUND), or the package is under 30 days old AND its name looks like a popular package's (PACKAGE_NEW_LOOKALIKE); otherwise "unknown" if any lookup did not complete (a failed registry lookup gives exists: null, never false); otherwise "no_flags". A look-alike name alone, a new package alone, known vulnerabilities, deprecation and install scripts are informational reasons that never flag on their own. This tool never returns "safe" or "clean"; no_flags means nothing was found in the sources checked. VALIDATION: one package per call; malformed input returns an MCP tool error (isError: true, code invalid_ecosystem, invalid_package_name or invalid_version). Registry answers are cached for 1 hour and malicious/vulnerability answers for 15 minutes. Rate-limited to 60 calls/hour per caller when unauthenticated; a valid Gumroad license key (X-API-KEY header) exempts the limit. Data sources and licences: https://pg1-ai-agent.vercel.app/docs/attributions
  • check_repo_trapPG1 Sovereign Threat Intelligence: checks a downloaded project BEFORE you run anything from it, for the untrusted repo trap: a project set up to hijack the AI coding agent that ope…3 findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.

    Findings

    • High

      The description refers to ".npmrc" and "~/.ssh".

      Matched text

      description: "...it for README, package.json, setup.py, .npmrc and the .py files at the root) }, up to..."

      What this check looks for
    • Info

      The description mentions "credentials".

      Matched text

      description: "...), SECRET_PATH_REFERENCE (~/.ssh, cloud credentials, GITHUB_TOKEN, keychains, internal addr..."

      What this check looks for
    • Info

      The description is 2,172 characters long, over the 2,000 this check looks for.

      Matched text

      description ends: "...: only the verdict and reason codes are logged. Always free (60/hour per caller)."

      What this check looks for
    Description, as the server wrote it
    PG1 Sovereign Threat Intelligence: checks a downloaded project BEFORE you run anything from it, for the untrusted repo trap: a project set up to hijack the AI coding agent that opens it. The classic trap ships a suspicious binary and an encoded data file; a careful agent refuses the binary and writes its own Python decoder, but a file named like a standard-library module (struct.py) sits in the working folder, Python puts that folder first on sys.path, and the agent's own script imports the attacker's file, which starts a detached process and takes SSH keys and tokens. Send the file listing you already have: files[] of { path, size?, executable?, magic_hex? (the first bytes as hex, up to 32), text? (the start of the file, up to 16 KB; send it for README, package.json, setup.py, .npmrc and the .py files at the root) }, up to 2000 files and 256 KB of text. PG1 never fetches a URL, unpacks, decodes, imports or runs anything from the project: deterministic rules over what you sent, no AI model. Reason codes: STDLIB_SHADOW and PY_STARTUP_HOOK (sitecustomize.py, usercustomize.py, .pth) are high; NODE_INSTALL_HOOK (install-time npm scripts, an unusual registry), UNTRUSTED_BINARY (ELF, Mach-O, PE, executable bit), ENCODED_PAYLOAD, DECODER_BAIT (a README telling the reader to decode, run or import something; text that only describes the risk is not flagged), SECRET_PATH_REFERENCE (~/.ssh, cloud credentials, GITHUB_TOKEN, keychains, internal addresses) and PERSISTENCE_HINT (nohup, setsid, launchctl, cron, systemd user units, double fork) are medium. Returns verdict malicious (a shadow or startup hook together with an encoded payload or a binary, or a shadow or hook file that itself reaches for secrets or detaches), suspicious (any other finding) or no_findings, each finding with the paths involved, and plain-language advice (do not run code from the folder; if you must decode, use python3 -I in a sandbox with no secrets and no network, after deleting or renaming the shadowing file). No findings does not prove the project is harmless; never "safe". Nothing is stored: only the verdict and reason codes are logged. Always free (60/hour per caller).
  • check_wallet_agePG1 Sovereign Threat Intelligence: reports when an EVM wallet address first appeared on a given chain, based on its earliest on-chain transfer history (in or out), plus whether the…No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    Description, as the server wrote it
    PG1 Sovereign Threat Intelligence: reports when an EVM wallet address first appeared on a given chain, based on its earliest on-chain transfer history (in or out), plus whether the address is a contract. No payment required — this tool is always free. SIBLING DIFFERENTIATION: Use for wallet age/history only. Do NOT use for sanctions screening (use check_wallet_sanctions), domain age (use check_domain_age), or hostname/phishing reputation (use check_hostname_reputation). BEHAVIOR: Returns { address, chain, found, first_seen, age_days, first_seen_block, first_direction, is_contract, note, source: "on-chain transfer history", cached }. A found:false result (with all other fields null except is_contract) means the address has no transfer history on that chain — a normal, common result for a brand-new or never-used address, not an error, and not evidence of legitimacy either way; this tool reports age and history only. Upstream lookup failures or timeouts return an MCP tool error (isError: true) instead of found:false, since a data-source outage must never be read as "brand-new wallet". VALIDATION: address must be a 0x-prefixed 40-hex-character EVM address (case-insensitive); chain, if given, must be one of the supported enum values. Malformed input returns an MCP tool error (isError: true, code invalid_address or invalid_chain) instead of a result. Rate-limited to 60 calls/hour per caller when unauthenticated; a valid Gumroad license key (X-API-KEY header) exempts the limit, same as check_domain_age. A found result is cached permanently (it never changes); a found:false result is cached for 10 minutes only, since a wallet can become active at any time.
  • check_wallet_sanctionsPG1 Sovereign Threat Intelligence: checks a cryptocurrency wallet address against the OFAC SDN (Specially Designated Nationals) sanctions list, synced daily from US Treasury data.No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    Description, as the server wrote it
    PG1 Sovereign Threat Intelligence: checks a cryptocurrency wallet address against the OFAC SDN (Specially Designated Nationals) sanctions list, synced daily from US Treasury data. No payment required — this tool is always free. SIBLING DIFFERENTIATION: Use for wallet/address sanctions screening only. Do NOT use for IP/domain/hash/URL threat lookups (use get_ioc_context) or CVE data (use get_cve_details). BEHAVIOR: Returns { listed: true|false, matches, source, list_last_synced }. A listed:false result means the address is not on the OFAC SDN list as of the reported sync time — it is informational only, not legal or sanctions-compliance advice, and is never phrased as "safe" or "clean". Fails loudly (returns an error) if the sanctions data is empty or unreachable, rather than ever reporting listed:false on a data failure. VALIDATION: if the address does not match a recognised format for any supported currency (EVM, BTC/LTC/BCH/DOGE/DASH/ZEC base58 or bech32/cashaddr, TRON, Monero, Solana), returns an MCP tool error (isError: true, code invalid_address) instead of a result — it never reports listed:false for malformed input.
  • check_x402_endpointPG1 Sovereign Threat Intelligence: checks an x402-paid URL before you pay it.Can send data outNo 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

    • outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.
    Description, as the server wrote it
    PG1 Sovereign Threat Intelligence: checks an x402-paid URL before you pay it. Fetches the URL once (https only; private addresses and redirects refused) and reads its x402 payment request: each option's network must be a known settlement network and its asset the known USDC contract there; the receiving wallet (payTo) is screened against the OFAC SDN list and its on-chain age read; TLS and the domain's age are checked. Always free (60/hour per caller). Returns status flagged / no_flags / unknown with reasons (X402_UNKNOWN_NETWORK, X402_UNKNOWN_ASSET, WALLET_SANCTIONED, WALLET_NEW_7D, ENDPOINT_TLS_FAILED, DOMAIN_NEWLY_REGISTERED_30D, X402_NO_PAYMENT_REQUEST). no_flags means no flags found, never "safe".
  • hosting_advisorPG1 hosting advisor: ranked regions and providers for a workload, with reasons: data-protection regime, published government data-access frameworks (factual, cited), provider accep…No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    Description, as the server wrote it
    PG1 hosting advisor: ranked regions and providers for a workload, with reasons: data-protection regime, published government data-access frameworks (factual, cited), provider acceptable-use fit, sanctions and export-control compliance, estimated latency, a multi-region/multi-provider redundancy and resilience plan, and privacy-by-design guidance. Every fact cites its official source and date; unverified facts and facts older than 180 days are flagged. Structured input only. Not legal advice. Lawful use only: requests to ignore abuse reports, evade lawful process or bypass sanctions are refused. Payment: $0.01 via x402 (params._meta["x402/payment"]) or a Gumroad licence key, after 5 free calls a day per caller shared with every other paid tool.
  • preflight_checkPG1 Sovereign Threat Intelligence: ONE call before an agent acts.No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    Description, as the server wrote it
    PG1 Sovereign Threat Intelligence: ONE call before an agent acts. Give the action you are about to take and its target; PG1 runs every relevant check in parallel and returns one decision: no_flags, flagged or unknown, with merged reasons, per-check results and latency. no_flags means no flags found in the checks run, never "safe". Fail closed: any check that does not finish makes the decision unknown, never no_flags. Actions: pay (a wallet address: sanctions + wallet age; or an https x402 URL: the endpoint's payTo sanctions and age, network/asset sanity, TLS, domain age), open (a URL or hostname: phishing/look-alike reputation + domain age), install (a package: give ecosystem npm or pypi; target name or name@version), patch (a CVE id, or "vendor product version"), trust_agent (an agent's URL or domain: its A2A agent card, domain age and reputation). Payment: $0.01 via x402 (params._meta["x402/payment"]) or a Gumroad licence key, after 5 free calls a day per caller shared with every other paid tool. Results are signed (see verify_result).
  • submit_indicatorPG1 Sovereign Threat Intelligence: submit an observed indicator for validation and possible inclusion in future query results.No findings

    Why it is in this group

    • The name "submit_indicator" says "submit": a word for a change.
    Description, as the server wrote it
    PG1 Sovereign Threat Intelligence: submit an observed indicator for validation and possible inclusion in future query results. Requires a valid Gumroad license key (X-API-KEY header) — this tool is NOT available via per-query x402. Submissions are staged for review, not immediately added to the live feed.
  • verify_resultPG1 Sovereign Threat Intelligence: verifies that a result really came from PG1, unchanged.No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    Description, as the server wrote it
    PG1 Sovereign Threat Intelligence: verifies that a result really came from PG1, unchanged. Every PG1 tool result is signed (Ed25519) over its request hash, result, sources, timestamp and PG1 version; the signature is in result._meta["pg1/signature"] (MCP), the task's metadata["pg1.signature"] (A2A) or "signature" (REST). Give the result object and the signature block; optionally tool and arguments to check it answers that exact request. Public keys: https://pg1-ai-agent.vercel.app/.well-known/pg1-signing-key (JWKS); you can verify offline without this tool. Always free.
Only reads 8Show 8 read-only tools
  • get_cve_batchPG1 Sovereign Threat Intelligence: looks up multiple CVE identifiers in a single call, each enriched with NVD description/CVSS and CISA KEV status (plus the FIRST.org EPSS score wh…No findings

    Why it is in this group

    • The name "get_cve_batch" says "get": a word for reading.
    Description, as the server wrote it
    PG1 Sovereign Threat Intelligence: looks up multiple CVE identifiers in a single call, each enriched with NVD description/CVSS and CISA KEV status (plus the FIRST.org EPSS score when this server serves EPSS in paid responses) — same enrichment as get_cve_details, batched. Payment required: $0.01 via x402, sent in params._meta['x402/payment'] (the PAYMENT-SIGNATURE header is also accepted), or a valid Gumroad license key (X-API-KEY header). SIBLING DIFFERENTIATION: Use for looking up several known CVE ids at once (e.g. from an SBOM or scan report). Do NOT use for a single CVE (use get_cve_details, lower overhead) or for discovering CVEs by vendor/product (use get_cve_by_product). BEHAVIOR: Accepts up to 20 ids per call; malformed or not-found ids are reported per-entry rather than failing the whole batch.
  • get_cve_by_productPG1 Sovereign Threat Intelligence: returns CVEs affecting a given vendor/product (optionally a specific version), enriched with CVSS and CISA KEV status (plus EPSS when this server…No findings

    Why it is in this group

    • The name "get_cve_by_product" says "get": a word for reading.
    Description, as the server wrote it
    PG1 Sovereign Threat Intelligence: returns CVEs affecting a given vendor/product (optionally a specific version), enriched with CVSS and CISA KEV status (plus EPSS when this server serves EPSS in paid responses), sorted by exploitation risk. Sourced from NVD keyword search. Payment required: $0.01 via x402, sent in params._meta["x402/payment"] (the PAYMENT-SIGNATURE header is also accepted), or a valid Gumroad license key (X-API-KEY header). SIBLING DIFFERENTIATION: Use for discovering CVEs by vendor/product when you do not already have an exact CVE id. Do NOT use for a known CVE id (use get_cve_details / get_cve_batch). USAGE EXCLUSIONS: Uses NVD keyword search, not strict CPE matching — results may include near-matches. BEHAVIOR: Returns up to 50 results per call.
  • get_cve_detailsPG1 Sovereign Threat Intelligence: enriched CVE lookup combining NVD (description, CVSS score/vector) and the CISA Known Exploited Vulnerabilities catalog (active wild exploitation…No findings

    Why it is in this group

    • The name "get_cve_details" says "get": a word for reading.
    Description, as the server wrote it
    PG1 Sovereign Threat Intelligence: enriched CVE lookup combining NVD (description, CVSS score/vector) and the CISA Known Exploited Vulnerabilities catalog (active wild exploitation status). The epss field (FIRST.org EPSS exploit-probability score and percentile) is present only when this server serves EPSS in paid responses. Payment required: $0.01 via x402, sent in params._meta["x402/payment"] (the PAYMENT-SIGNATURE header is also accepted), or a valid Gumroad license key (X-API-KEY header). SIBLING DIFFERENTIATION: Use ONLY for specific CVE lookups. Do NOT use for IP/domain/hash enrichment (use get_ioc_context) or bulk feed ingestion (use get_threat_indicators). USAGE EXCLUSIONS: Does not support wildcard search or threat-actor dossier profiling. BEHAVIOR: If payment is missing or fails, returns a normal tool result with isError: true, the x402 v2 PaymentRequired object in structuredContent and the same JSON in content[0].text; on success the settlement receipt is in result._meta["x402/payment-response"]. Returns 404 if CVE is not found.
  • get_ioc_batchPG1 Sovereign Threat Intelligence: looks up multiple indicators (IPs, domains, URLs, hashes) in a single call — a batched pre-action safety check for AI agents.No findings

    Why it is in this group

    • The name "get_ioc_batch" says "get": a word for reading.
    Description, as the server wrote it
    PG1 Sovereign Threat Intelligence: looks up multiple indicators (IPs, domains, URLs, hashes) in a single call — a batched pre-action safety check for AI agents. Each returns the same aggregated provenance as get_ioc_context from PG1's commercially licensed records, which currently hold only CVE identifiers from NVD (IPs, domains, URLs and hashes return found: false). SIBLING DIFFERENTIATION: Use for checking several indicators at once (e.g. all URLs an agent is about to visit). Do NOT use for a single indicator (use get_ioc_context, lower overhead) or bulk feed synchronization (use get_threat_indicators). BEHAVIOR: Accepts up to 20 indicators per call. A found:false result for any indicator means nothing bad is recorded in PG1's sources — NOT that it's safe. If NONE of the submitted indicators are found, the whole batch is FREE — no payment or free-tier quota consumed. If at least one indicator is found, the normal payment gate (x402 via params._meta['x402/payment'], with the PAYMENT-SIGNATURE header also accepted, or a Gumroad X-API-KEY license) applies to the full batch result.
  • get_ioc_contextPG1 Sovereign Threat Intelligence: looks up a single specific indicator value (IP, domain, URL, or hash) — the recommended pre-action safety check for AI agents before visiting, do…Reads text others writeNo findings

    Why it is in this group

    • The name "get_ioc_context" says "get": a word for reading.

    What it can reach

    • untrusted-inputThe description's first sentence says "url": it reads web pages, which anyone can write.
    Description, as the server wrote it
    PG1 Sovereign Threat Intelligence: looks up a single specific indicator value (IP, domain, URL, or hash) — the recommended pre-action safety check for AI agents before visiting, downloading, or connecting to something. PG1's commercially licensed records currently hold only CVE identifiers from NVD, so an IP, domain, URL or hash returns found: false (free) until other licensed sources are added. A match returns aggregated provenance — reporting sources, observation count, aggregated confidence score, known malware families, tags, and first/last seen timestamps. SIBLING DIFFERENTIATION: Use ONLY for point-lookup enrichment of a single indicator. Do NOT use for bulk intelligence downloads (use get_threat_indicators), multiple indicators at once (use get_ioc_batch), or software vulnerability analysis (use get_cve_details). BEHAVIOR: Returns a normal result shaped { found: true, indicator_type, provenance } or { found: false } — never an error for 'not found'. A found:false result means nothing bad is recorded in PG1's sources; it does NOT mean the indicator is safe, only that it isn't in this dataset. Lookups that return found:false are FREE — no payment or free-tier quota is consumed. Payment (x402 via params._meta['x402/payment'], with the PAYMENT-SIGNATURE header also accepted, or a Gumroad X-API-KEY license) is only required when a real record is found. If payment is required but missing or fails, the result has isError: true with the x402 v2 PaymentRequired object in structuredContent.
  • get_threat_actor_profilePG1 Sovereign Threat Intelligence: returns a dossier for a known threat actor / APT group — aliases, description, associated MITRE ATT&CK techniques, and associated malware/tooling…No findings

    Why it is in this group

    • The name "get_threat_actor_profile" says "get": a word for reading.
    Description, as the server wrote it
    PG1 Sovereign Threat Intelligence: returns a dossier for a known threat actor / APT group — aliases, description, associated MITRE ATT&CK techniques, and associated malware/tooling. Sourced from MITRE ATT&CK Enterprise. Payment required: $0.01 via x402, sent in params._meta['x402/payment'] (the PAYMENT-SIGNATURE header is also accepted), or a valid Gumroad license key (X-API-KEY header). SIBLING DIFFERENTIATION: Use for actor/group-level profiling. Do NOT use for single-indicator lookups (use get_ioc_context) or vulnerability data (use get_cve_details / get_cve_batch). USAGE EXCLUSIONS: Coverage is limited to groups tracked in MITRE ATT&CK — not all threat actors have an entry. BEHAVIOR: Returns 404 if no matching group or alias is found.
  • get_threat_indicatorsPG1 Sovereign Threat Intelligence: returns a STIX 2.1 bundle of CVE records from NIST NVD, the only source PG1 may currently redistribute commercially.No findings

    Why it is in this group

    • The name "get_threat_indicators" says "get": a word for reading.
    Description, as the server wrote it
    PG1 Sovereign Threat Intelligence: returns a STIX 2.1 bundle of CVE records from NIST NVD, the only source PG1 may currently redistribute commercially. Today it contains no IP, domain, URL or file-hash indicators: every object is a STIX vulnerability with the CVE id and an NVD reference, without scores (use get_cve_details for CVSS and known-exploited status). FREE (CVE records from NIST NVD); does not use the 5 free paid calls a day; no payment, no licence key (the operator's PG1_FEED_PAID switch can make it paid). SIBLING DIFFERENTIATION: Use ONLY for bulk feed synchronizations. Do NOT use for single-item lookups (use get_ioc_context) or CVE analysis (use get_cve_details). USAGE EXCLUSIONS: Does not provide historical query archival beyond the active ingestion window. BEHAVIOR: Pagination is handled via the limit parameter (max 1000).
  • get_usage_statusPG1 Sovereign Threat Intelligence: returns your remaining free-tier calls for today and current Gumroad license status.No findings

    Why it is in this group

    • The name "get_usage_status" says "get": a word for reading.
    Description, as the server wrote it
    PG1 Sovereign Threat Intelligence: returns your remaining free-tier calls for today and current Gumroad license status. No payment required — this tool is always free.

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 private-data.

Reads private data0
Reads files, mail, messages, databases, secrets, calendars or repositories.
Reads text others write1
Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
Can send data out3
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.

Sign-in required, no OAuth metadata

Looked at Nov 20, 2026, 19:00 UTC, unchanged since Nov 13, 2026. No credentials were sent and no tool was called. What is looked at.

Reachable
yes
HTTP status of an anonymous initialize
402
tools/list served without credentials
no
Tools that change things, listed without credentials
0
Protocol version
not observed
WWW-Authenticate on 401
not observed
resource_metadata in WWW-Authenticate
not observed
RFC 9728 protected resource metadata
not observed
Authorization servers
not observed
Legacy HTTP+SSE transport
no
Dynamic client registration
not observed
Client ID metadata documents
not observed

The probe stopped: initialize answered HTTP 402.

Posture changes

  1. Nov 13, 2026
    • HTTP status of an anonymous initialize: 200 to 402
    • tools/list served without credentials: yes to no
    • Tools that change things, listed without credentials: 13 to 0
    • Protocol version: 2025-06-18 to none
  2. Oct 29, 2026
    • Tools that change things, listed without credentials: 12 to 13

Findings

Matches of fixed checks, with the text they matched.

3 high5 info
  • The description contains "ignore previous instructions", an instruction to set earlier instructions aside.

    Matched text

    description: "...: instruction override and role hijack (ignore previous instructions, "you are now", developer or jailbreak..."

    What this check looks for
  • The description refers to ".env" and "~/.ssh".

    Matched text

    description: "...ironment variables, keys, seed phrases, .env or ~/.ssh; image or link URLs carrying..."

    What this check looks for
  • The description refers to ".npmrc" and "~/.ssh".

    Matched text

    description: "...it for README, package.json, setup.py, .npmrc and the .py files at the root) }, up to..."

    What this check looks for
  • The description is 2,331 characters long, over the 2,000 this check looks for.

    Matched text

    description ends: "...gth, kind, verdict and reason codes are logged. Always free (60/hour per caller)."

    What this check looks for
  • The description mentions "credentials".

    Matched text

    description: "...), SECRET_PATH_REFERENCE (~/.ssh, cloud credentials, GITHUB_TOKEN, keychains, internal addr..."

    What this check looks for
  • The description is 2,172 characters long, over the 2,000 this check looks for.

    Matched text

    description ends: "...: only the verdict and reason codes are logged. Always free (60/hour per caller)."

    What this check looks for
Show 2 more findings
  • The description is 2,321 characters long, over the 2,000 this check looks for.

    Matched text

    description ends: "...nauthenticated; a valid Gumroad license key (X-API-KEY header) exempts the limit."

    What this check looks for
  • The description is 2,505 characters long, over the 2,000 this check looks for.

    Matched text

    description ends: "...mit. Data sources and licences: https://pg1-ai-agent.vercel.app/docs/attributions"

    What this check looks for

History

Changes to its tool definitions and to its remote endpoint, newest first.

  1. Tools1 added.+1 addedView diff
  2. PostureRemote posture changed. HTTP status of an anonymous initialize: 200 -> 402; tools/list served without credentials: yes -> no; Tools that change things, listed without credentials: 13 -> 0 (and 1 more).View diff
    • HTTP status of an anonymous initialize: 200 to 402
    • tools/list served without credentials: yes to no
    • Tools that change things, listed without credentials: 13 to 0
    • Protocol version: 2025-06-18 to none
  3. PostureRemote posture changed. Tools that change things, listed without credentials: 12 -> 13.View diff
    • Tools that change things, listed without credentials: 12 to 13
  4. Tools1 added.+1 addedView diff

Install

Pinned to version 1.22.0. Values in angle brackets are yours to fill in.

claude mcp add --transport http pg1-threat-intel https://pg1-ai-agent.vercel.app/api/mcp
claude mcp add --transport http pg1-threat-intel https://pg1-ai-agent.vercel.app/api/mcp

Run 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.

.claude/settings.json
{
  "permissions": {
    "allow": [
      "mcp__pg1-threat-intel__get_threat_indicators",
      "mcp__pg1-threat-intel__get_cve_details",
      "mcp__pg1-threat-intel__get_ioc_context",
      "mcp__pg1-threat-intel__get_cve_batch",
      "mcp__pg1-threat-intel__get_ioc_batch",
      "mcp__pg1-threat-intel__get_threat_actor_profile",
      "mcp__pg1-threat-intel__get_cve_by_product",
      "mcp__pg1-threat-intel__get_usage_status"
    ],
    "ask": [
      "mcp__pg1-threat-intel__preflight_check",
      "mcp__pg1-threat-intel__check_x402_endpoint",
      "mcp__pg1-threat-intel__check_agent_card",
      "mcp__pg1-threat-intel__check_agent_message",
      "mcp__pg1-threat-intel__check_repo_trap",
      "mcp__pg1-threat-intel__verify_result",
      "mcp__pg1-threat-intel__hosting_advisor",
      "mcp__pg1-threat-intel__submit_indicator",
      "mcp__pg1-threat-intel__check_wallet_sanctions",
      "mcp__pg1-threat-intel__check_domain_age",
      "mcp__pg1-threat-intel__check_hostname_reputation",
      "mcp__pg1-threat-intel__check_wallet_age",
      "mcp__pg1-threat-intel__check_ip_abuse",
      "mcp__pg1-threat-intel__check_package"
    ],
    "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 pg1-threat-intel. Review it before you use it.

Capability tags behind 4 rules
  • mcp__pg1-threat-intel__check_x402_endpoint outbound
  • mcp__pg1-threat-intel__check_agent_card outbound
  • mcp__pg1-threat-intel__check_agent_message outbound
  • mcp__pg1-threat-intel__get_ioc_context 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.Project-Gifted1/pg1-threat-intel/verify?hash=7fb4856718f4769ab81b2956033dc8d382b6f68f10f546dcceb1f488791d11d2" | jq -e '.changed == false'

Pinned hash 7fb4856718f4, from Nov 21, 2026, 03:18 UTC.

Watch

Get told when its tools change: a new tool, a new description, a new class.

https://radar.anlyon.com/servers/io.github.Project-Gifted1/pg1-threat-intel/changes.xml

An Atom feed. Works in any feed reader or Slack's RSS app.

Badge

Anlyon Radar badge for io.github.Project-Gifted1/pg1-threat-intel
[![Anlyon Radar](https://radar.anlyon.com/badge/io.github.Project-Gifted1/pg1-threat-intel.svg)](https://radar.anlyon.com/servers/io.github.Project-Gifted1/pg1-threat-intel)

Source: MCP Registry · repository. Synced Nov 21, 2026, 03:17 UTC.

PG1 Sovereign Threat Intelligence MCP server: questions and answers

Answered from the scan above, not from the publisher's description.

Is the PG1 Sovereign Threat Intelligence MCP server safe?

No scan can promise that. What its definitions show: 22 tools, 14 of which can change, send or delete something, and 8 findings (3 high) from fixed checks. Allow the reads, ask before each change, and deny the deletes; the permissions block on this page does that.

Can the PG1 Sovereign Threat Intelligence MCP server delete data?

No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.

Can the PG1 Sovereign Threat Intelligence 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 PG1 Sovereign Threat Intelligence MCP server have?

22 tools: 14 write, 8 read. preflight_check, check_x402_endpoint, check_agent_card, check_agent_message, check_repo_trap, verify_result, hosting_advisor, submit_indicator and 14 more.

Does the PG1 Sovereign Threat Intelligence MCP server need sign-in or OAuth?

Its remote endpoint did not answer the last anonymous look.

Does the PG1 Sovereign Threat Intelligence MCP server send data to outside services?

3 tools can carry data out (it sends, posts or publishes, or takes a recipient, channel or URL): check_x402_endpoint, check_agent_card and check_agent_message.

Has the PG1 Sovereign Threat Intelligence MCP server changed its tools?

Yes, 2 times since Anlyon Radar started tracking it, most recently on Nov 21, 2026. Every change is on this page and in its Atom feed.