cap-shield MCP server
io.github.robinlidberg-dot/cap-shieldrobinlidberg-dotstreamable-httpv0.7.0Scanned
Context selection and compression for AI agents - recall measured, not claimed.
9 tools: 6 change things and 3 only read.
- Reads private data · 0
- Reads text others write · 0
- Can send data out · 0
- 2 findings
Tools listed without sign-in. Remote posture
With Anlyon
6 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 cap-shield MCP server can do
The cap-shield MCP server is published by the GitHub account robinlidberg-dot. It has 9 tools. It can create or change data (6 tools: measure_traffic, remember, get_account, issue_pass and 2 more), but cannot delete data. 3 tools only read.
None of its tools reads private data, brings in text other people write, or sends data out.
The fixed checks matched 2 findings (annotation-conflict and credentials-mention). Each one comes with the text it matched.
Its remote endpoint lists its tools without sign-in.
Last scanned Nov 17, 2026, version 0.7.0.
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.
9 tools
Changes things 6Claude Code: ask
get_accountGet an account and an API key.1 finding
Why it is in this group
- The name "get_account" says "get": a word for reading.
- The server marks it as not read-only (readOnlyHint: false).
Findings
- Warn
The server marks "get_account" as not read-only (readOnlyHint: false), and its words say it reads.
What this check looks forMatched text
annotations: {"readOnlyHint":false,"destructiveHint":false,"idempotentHint":false,"openWorldHint":false}; The name "get_account" says "get": a word for reading.
Description, as the server wrote it Get an account and an API key. Requires an email address. The key is returned ONCE and cannot be shown again — store it immediately. Beta quotas are low by design; they are hard stops, never overage billing.
Annotations the server declares
- title: Create account and API key
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: false
issue_passIssue a short-lived (15 minute) signed credential (a JWT) stating who the caller acts for.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 mentions "credential".
What this check looks forMatched text
description: "Issue a short-lived (15 minute) signed credential (a JWT) stating who the caller acts for..."
Description, as the server wrote it Issue a short-lived (15 minute) signed credential (a JWT) stating who the caller acts for. REQUIRES A KEY. Anyone can verify it independently against /.well-known/jwks.json — without calling CAP-Shield again. 'acts_for' and 'scope' are NOT validated against reality: the pass only attests that the key holder claimed this, at this time, and that the claim is recorded in the audit chain. This is identity and claimed authorization, not a payment or access-control mechanism.
Annotations the server declares
- title: Issue a signed agent pass
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: false
measure_trafficMeasure how much of your own agent traffic could be saved.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.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it Measure how much of your own agent traffic could be saved. NO ACCOUNT OR KEY NEEDED — use this first. Returns byte savings over the wire and, if a query is given, token savings from selective context retrieval. Nothing is stored: the text is compressed in memory and discarded. Rate limited to 20 calls per hour per IP.
Annotations the server declares
- title: Measure possible savings
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
pack_batchPack several messages into ONE compressed packet instead of packing them one at a time.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.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it Pack several messages into ONE compressed packet instead of packing them one at a time. REQUIRES A KEY. Worth it for many small/medium messages (the per-message header cost is paid once per batch, and zstd shares context across them) — for a single message this has no advantage. Returns packet_b64 (hand it to unpack_batch to get the messages back) plus the measured byte savings versus packing each message separately. SECURITY: only batch messages that already share the SAME trust boundary (same tenant/agent — nothing attacker-influenced mixed in with a secret). Batched messages share one compression context, which can leak size-based information between them (a CRIME/BREACH-style side channel). There is no padding mitigation on this endpoint today, so trust-boundary separation is the only protection — don't rely on padding.
Annotations the server declares
- title: Compress a batch of messages
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
rememberStore a memory entry for later retrieval.No findings
Why it is in this group
- The description's first verb is "store": a word for a change.
Description, as the server wrote it Store a memory entry for later retrieval. REQUIRES A KEY. This does not call any language model — it stores text in an isolated per-tenant archive. Use assemble_context to get relevant entries back.
Annotations the server declares
- title: Store a memory entry
- readOnlyHint: false
- destructiveHint: false
- idempotentHint: false
- openWorldHint: false
unpack_batchUnpack a batch packet made by pack_batch (or by the /api/v1/pack-batch endpoint) back into the exact original list of messages, in order.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.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Description, as the server wrote it Unpack a batch packet made by pack_batch (or by the /api/v1/pack-batch endpoint) back into the exact original list of messages, in order. REQUIRES A KEY. Not for a packet made by the single-message pack endpoint — that is a different format and this will reject it.
Annotations the server declares
- title: Decompress a batch packet
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
Only reads 3Show 3 read-only tools
assemble_contextRetrieve the memory entries that answer a question, within a token budget.No findings
Why it is in this group
- The description's first verb is "retrieve": a word for reading.
Description, as the server wrote it Retrieve the memory entries that answer a question, within a token budget. REQUIRES A KEY. Send the returned 'context' to your language model INSTEAD of the whole history. This does not call a model itself — it selects what to send. The budget is a ceiling, not a target: selection stops where relevance runs out, often well below it. The response says how many entries were left behind and why. If the response has a 'budget_warning', the best match did not fit: call again with its suggested_token_budget.
Annotations the server declares
- title: Retrieve relevant memory
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
assemble_escalation_contextRetrieve a WIDE context for a question your model could not answer from assemble_context.No findings
Why it is in this group
- The description's first verb is "retrieve": a word for reading.
Description, as the server wrote it Retrieve a WIDE context for a question your model could not answer from assemble_context. REQUIRES A KEY. Use it when assemble_context returned escalate_recommended=true, or when the answer depends on dates, counts or several sessions. This does not call a model itself. Returns 'context' (up to token_budget, default 150 000 tokens: lower it if your model's window is smaller), where each stored entry starts with a header '--- Session YYYY/MM/DD (Ddd) HH:MM ---' giving when it was stored, and 'utelamnat_kontext': lower-ranked entries that were left out, in the same format. Read those only if the answer is still missing from 'context'. Python users can pass the whole response to cap_escalate_helper.eskalera_om_behovs_v6().
Annotations the server declares
- title: Retrieve wide dated memory
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
list_packagesList the available dictionaries with their MEASURED compression, including the ones that perform badly.No findings
Why it is in this group
- The name "list_packages" says "list": a word for reading.
Description, as the server wrote it List the available dictionaries with their MEASURED compression, including the ones that perform badly. Each entry says whether it works one message at a time or only batched, and how many messages came out LARGER. No key needed.
Annotations the server declares
- title: List compression packages
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
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 or untrusted-input or outbound.
- Reads private data0
- Reads files, mail, messages, databases, secrets, calendars or repositories.
- Reads text others write0
- 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 18, 2026, 16:20 UTC, unchanged since Nov 18, 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
- 6
- Protocol version
- 2024-11-05
- 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
- WarnThe whole server
The endpoint served tools/list without credentials, and the list has 6 tools that change things (6 write). Whether a call also works without credentials was not tested: tools are never called.
What this check looks forMatched text
https://cap-shield-robin.fly.dev/mcp: tools/list returned 9 tools to an anonymous client, including measure_traffic (write), remember (write), get_account (write), issue_pass (write), pack_batch (write) and 1 more
Findings
Matches of fixed checks, with the text they matched.
- Warnget_account
The server marks "get_account" as not read-only (readOnlyHint: false), and its words say it reads.
What this check looks forMatched text
annotations: {"readOnlyHint":false,"destructiveHint":false,"idempotentHint":false,"openWorldHint":false}; The name "get_account" says "get": a word for reading.
- Infoissue_pass
The description mentions "credential".
What this check looks forMatched text
description: "Issue a short-lived (15 minute) signed credential (a JWT) stating who the caller acts for..."
History
Changes to its tool definitions and to its remote endpoint, newest first.
No change seen yet. Its tools are rescanned every day, and a remote endpoint is looked at once a week.
Install
Pinned to version 0.7.0. Values in angle brackets are yours to fill in.
claude mcp add --transport http cap-shield https://cap-shield-robin.fly.dev/mcpclaude mcp add --transport http cap-shield https://cap-shield-robin.fly.dev/mcpRun 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__cap-shield__list_packages",
"mcp__cap-shield__assemble_context",
"mcp__cap-shield__assemble_escalation_context"
],
"ask": [
"mcp__cap-shield__measure_traffic",
"mcp__cap-shield__remember",
"mcp__cap-shield__get_account",
"mcp__cap-shield__issue_pass",
"mcp__cap-shield__pack_batch",
"mcp__cap-shield__unpack_batch"
],
"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 cap-shield. Review it before you use it.
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.robinlidberg-dot/cap-shield/verify?hash=b6f70f5bac13ccf38a81228bf6b7ab9b7c8824aeacaf8b9dbee0f99dec3537c9" | jq -e '.changed == false'Pinned hash b6f70f5bac13, from Nov 17, 2026, 22:18 UTC.
Watch
Get told when its tools change: a new tool, a new description, a new class.
https://radar.anlyon.com/servers/io.github.robinlidberg-dot/cap-shield/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://radar.anlyon.com/servers/io.github.robinlidberg-dot/cap-shield)Source: MCP Registry · repository. Synced Nov 17, 2026, 22:18 UTC.
cap-shield MCP server: questions and answers
Answered from the scan above, not from the publisher's description.
Is the cap-shield MCP server safe?
- No scan can promise that. What its definitions show: 9 tools, 6 of which can change, send or delete something, and 2 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 cap-shield MCP server delete data?
- No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.
Can the cap-shield 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 cap-shield MCP server have?
- 9 tools: 6 write, 3 read. measure_traffic, remember, get_account, issue_pass, pack_batch, unpack_batch, list_packages, assemble_context and 1 more.
Does the cap-shield MCP server need sign-in or OAuth?
- Its remote endpoint listed its tools to an anonymous client, without sign-in.
Does the cap-shield MCP server send data to outside services?
- None of its tools sends, posts or publishes, or takes a recipient, channel or URL.
Has the cap-shield MCP server changed its tools?
- Not since Anlyon Radar started tracking it. Last scanned Nov 17, 2026.