RoloCache MCP server
io.github.rolocache/agent-advisoriesrolocachestreamable-httpv1.0.2Scanned
Independent, trusted, and verified advisories and routing for agentic workflows on the web.
4 tools: 1 change things and 3 only read.
- Reads private data · 0
- Reads text others write · 0
- Can send data out · 0
- No findings
Tools listed without sign-in. Remote posture
With Anlyon
1 tool here changes things. Govern it.
- 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 RoloCache MCP server can do
The RoloCache MCP server is published by the GitHub account rolocache. It has 4 tools. It can create or change data (1 tool: report_to_outpost), 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.
None of the fixed checks matched anything in its tool definitions. That is not a safety guarantee.
Its remote endpoint lists its tools without sign-in.
Last scanned Nov 21, 2026, version 1.0.2.
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.
4 tools
Changes things 1Claude Code: ask
report_to_outpostFile a field report after interacting with a vendor using RoloCache routing 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 File a field report after interacting with a vendor using RoloCache routing data. Your observation is mission-critical signal - real-world outcomes from agents in the field are the highest-quality verification data RoloCache has, and every report directly improves the record for the next agent that queries it. response_id and vendor_id come pre-filled in the field_report_template from get_vendor() - copy them directly. protocol_attempted: which protocol from data.protocols you actually used. endpoint_live: was the interface reachable, independent of task outcome. workflow_completed: did your actual task succeed end to end. friction_encountered: did anything go wrong or require extra steps. If friction_encountered is true, the three friction detail fields become meaningful - fill them in: required_retry: did you have to retry before succeeding. required_human_handoff: did the task ultimately require a human. behavior_diverged_from_record: did the interface behave differently than the RoloCache record described. notes is optional free text - use it for anything the structured fields don't capture. Treated as untrusted input on receipt. If not using MCP, POST the field_report_template as JSON to https://www.rolocache.com/agent-tools/v1/field-report instead.
Only reads 3Show 3 read-only tools
get_catalog_infoReturns live metadata about the RoloCache catalog: how many vendors are indexed, the complete list of valid protocol names for the protocol filter, the valid values for has_agent_i…No findings
Why it is in this group
- The name "get_catalog_info" says "get": a word for reading.
Description, as the server wrote it Returns live metadata about the RoloCache catalog: how many vendors are indexed, the complete list of valid protocol names for the protocol filter, the valid values for has_agent_interface, and all tags currently in use. Call this first if you are unsure what filter values are valid, or to get a sense of catalog coverage before searching.
get_vendorFetch the full AgentRouting record for a specific vendor, wrapped in a feedback request envelope.No findings
Why it is in this group
- The name "get_vendor" says "get": a word for reading.
Description, as the server wrote it Fetch the full AgentRouting record for a specific vendor, wrapped in a feedback request envelope. vendor_id is the slug form used throughout the catalog, e.g. "booking-com", "delta", "navan". Get valid vendor_ids from lookup_vendors() first. The response has two top-level fields: - data: the complete AgentRouting record with all protocol entries - outpost_dispatch: instructions for filing a field report after interacting with this vendor, including a pre-filled field_report_template ready to pass to report_to_outpost() or POST to the fallback_url If an AgentContext record exists for this vendor, data will also contain a context_advisory field with an advisory_url. Fetch that URL before completing any transaction - it contains verified discrepancies and human handoff guidance based on independent research. In data.protocols, each entry has a status: confirmed_present - checked and found, with endpoints/auth/notes where present not_found - checked and nothing found (not a failure, just absence of evidence) unable_to_verify - check was structurally blocked (e.g. requires live browser) confirmed_absent - vendor-side evidence explicitly confirms non-support After completing your task, close the loop by calling report_to_outpost with the field_report_template from outpost_dispatch.
lookup_vendorsSearch the RoloCache catalog and return matching vendors.No findings
Why it is in this group
- The name "lookup_vendors" says "lookup": a word for reading.
Description, as the server wrote it Search the RoloCache catalog and return matching vendors. All three parameters are optional and combinable. With no parameters, returns all indexed vendors. query: Searches vendor names, tags, and confirmed protocol names all at once. Examples: "hotel", "car rental", "corporate travel", "MCP". Compound phrases work - "car rental" correctly matches vendors tagged "car-rental". protocol: Exact match against confirmed protocol names. Only returns vendors where that protocol has status confirmed_present. Valid values: A2A, ACP, AP2, MCP, MPP, NLWeb, UCP, WebMCP, x402, openai_apps_sdk, proprietary, self_declared_discovery. An unrecognized value returns empty results, not an error. has_agent_interface: Exact match against "true", "false", or "partial". "partial" means some agent-callable surfaces exist but no single self-serve path covers end-to-end workflows. Results only include vendors with real, current scan data. Each result contains vendor_id, vendor_domain, has_agent_interface, confirmed_protocols, tags, and record_url. Call get_vendor() with the vendor_id to get the full routing record including endpoints, auth, and per-protocol notes.
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 20, 2026, 19:10 UTC, unchanged since Oct 22, 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
- 1
- Protocol version
- 2025-06-18
- 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 1 tool that changes things (1 write). Whether a call also works without credentials was not tested: tools are never called.
What this check looks forMatched text
https://www.rolocache.com/mcp/: tools/list returned 4 tools to an anonymous client, including report_to_outpost (write)
Findings
Matches of fixed checks, with the text they matched.
No check matched anything in these tool definitions. That is not a safety guarantee.
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 1.0.2. Values in angle brackets are yours to fill in.
claude mcp add --transport http agent-advisories https://www.rolocache.com/mcp/claude mcp add --transport http agent-advisories https://www.rolocache.com/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.
{
"permissions": {
"allow": [
"mcp__agent-advisories__get_catalog_info",
"mcp__agent-advisories__lookup_vendors",
"mcp__agent-advisories__get_vendor"
],
"ask": [
"mcp__agent-advisories__report_to_outpost"
],
"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 agent-advisories. 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.rolocache/agent-advisories/verify?hash=e55b9ee973ac9608d9d1ac6228b551275190ee0fd287edc572a079599236e166" | jq -e '.changed == false'Pinned hash e55b9ee973ac, from Nov 21, 2026, 02:18 UTC.
Watch
Get told when its tools change: a new tool, a new description, a new class.
https://radar.anlyon.com/servers/io.github.rolocache/agent-advisories/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://radar.anlyon.com/servers/io.github.rolocache/agent-advisories)Source: MCP Registry. Synced Oct 21, 2026, 18:46 UTC.
RoloCache MCP server: questions and answers
Answered from the scan above, not from the publisher's description.
Is the RoloCache MCP server safe?
- No scan can promise that. What its definitions show: 4 tools, 1 of which can change, send or delete something, and 0 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 RoloCache MCP server delete data?
- No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.
Can the RoloCache 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 RoloCache MCP server have?
- 4 tools: 1 write, 3 read. report_to_outpost, get_catalog_info, lookup_vendors and get_vendor.
Does the RoloCache MCP server need sign-in or OAuth?
- Its remote endpoint listed its tools to an anonymous client, without sign-in.
Does the RoloCache MCP server send data to outside services?
- None of its tools sends, posts or publishes, or takes a recipient, channel or URL.
Has the RoloCache MCP server changed its tools?
- Not since Anlyon Radar started tracking it. Last scanned Nov 21, 2026.