Palimpsest — censorship, China economy and model-eval observatory MCP server
io.github.beepboop2025/palimpsestbeepboop2025streamable-httpv1.9.3Scanned
Rights-aware censorship, China-economic status, and tamper-evident AI evaluation tools.
7 tools: 2 change things and 5 only read.
- Reads private data · 0
- Reads text others write · 2
- Can send data out · 0
- No findings
Tools listed without sign-in. Remote posture
With Anlyon
2 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 Palimpsest — censorship, China economy and model-eval observatory MCP server can do
The Palimpsest — censorship, China economy and model-eval observatory MCP server is published by the GitHub account beepboop2025. It has 7 tools. It can create or change data (2 tools: research_catalog and whats_happening), but cannot delete data. 5 tools only read.
2 tools bring in text other people can write (list_signals and gfw_reading).
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, and its tools changed 2 times since Anlyon Radar first saw them.
Last scanned Nov 17, 2026, version 1.9.3.
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.
7 tools
Changes things 2Claude Code: ask
research_catalogDiscover every Palimpsest dataset, including the latest China economic histories, external trade, regional research, information controls and model evaluations.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 Discover every Palimpsest dataset, including the latest China economic histories, external trade, regional research, information controls and model evaluations. Paginated metadata retains source clocks, rights and access states. Includes the Seiche research_network handoff for funding, institution and market-liquidity context; values are not copied.
Annotations the server declares
- title: Complete research catalog and Seiche connection
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
whats_happeningJudge whether anything is happening in Chinese censorship right now, across every signal at once: the board's own cross-signal verdict with the multiplicity paid for (false-discove…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 Judge whether anything is happening in Chinese censorship right now, across every signal at once: the board's own cross-signal verdict with the multiplicity paid for (false-discovery control) and coverage confounds flagged as measurement artifacts, never findings. Takes no arguments. Use this instead of fetching signals individually and reconciling them yourself; then use get_signal to drill into whichever signal moved. Scope note: this is the censorship board. For the AI-model-evaluation side use get_signal with 'eval-registry', 'eval-assurance' or 'refusal-drift'.
Annotations the server declares
- title: Cross-signal board verdict
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
Only reads 5Show 5 read-only tools
get_newsroomRead one evidence-first reporting surface without scraping a page or guessing a filename.No findings
Why it is in this group
- The name "get_newsroom" says "get": a word for reading.
Description, as the server wrote it Read one evidence-first reporting surface without scraping a page or guessing a filename. Views: 'newsroom' for prioritized deterministic stories, 'wire' for normalized source dossiers, 'economy' for the currently restricted China economic pulse, 'machine-analysis' for the currently restricted AnalysisReports and AbstentionReports, 'investigations' for review-gated research leads, 'editorial-readiness' for publication gates, and 'interconnection' for named-key fat-object joins on China situation events (topic-surface-only, never wire corroboration). The economy, machine-analysis and interconnection views are metadata-only until a lineage-filtered rebuild removes denied derivatives. Availability never implies publication readiness: statuses, gates, counterevidence, limitations and right-to-reply state stay attached.
Annotations the server declares
- title: Evidence and reporting desks
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
get_signalRead one named signal.No findings
Why it is in this group
- The name "get_signal" says "get": a word for reading.
Description, as the server wrote it Read one named signal. Permitted signals return their bounded latest payload; China-economic signals in the denied lineage closure return explicit restricted/unavailable rights metadata and no values. Call list_signals first to discover valid names. Use this for the AI-model-evaluation side too: 'eval-registry' returns the pre-registered, hash-chained eval ledger with its verified flag and Merkle root, 'gfi-transcripts' returns a bounded view of the complete GFI v2 response matrix, and 'refusal-drift' returns the current frontier-model refusal reading on the frozen benign probe set; read 'eval-assurance' before turning either into a validity claim, and 'eval-journal' for the evidence-bound explanation and source receipts, or 'eval-findings' for the current deterministic article edition. Distinct from gfw_reading, which merges the two Great Firewall layers into one combined view.
Annotations the server declares
- title: One signal's full reading
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
gfw_readingRead the Great Firewall's current state at both layers in one call: live network blocking measured inside China via OONI (website, messenger and circumvention-tool reachability) jo…Reads text others writeNo findings
Why it is in this group
- The description's first verb is "read": a word for reading.
What it can reach
- untrusted-inputThe description's first sentence says "read" and "website": it reads web pages, which anyone can write.
Description, as the server wrote it Read the Great Firewall's current state at both layers in one call: live network blocking measured inside China via OONI (website, messenger and circumvention-tool reachability) joined with model-layer censorship from the Generative Firewall Index over Chinese LLMs. Takes no arguments. A combined convenience view. For one layer's full raw payload use get_signal with 'ooni-gfw' or 'generative-firewall-index'.
Annotations the server declares
- title: Great Firewall: both layers
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
inspect_economic_publication_rightsInspect publication-rights metadata for China economic evidence.No findings
Why it is in this group
- The name "inspect_economic_publication_rights" says "inspect": a word for reading.
Description, as the server wrote it Inspect publication-rights metadata for China economic evidence. Returns the policy digest, UTC evaluation clocks, per-source decisions and published-record status. Takes no arguments and never reads or returns observation rows, derived signals or replacement values. A denied publication decision remains a restriction, not a zero observation.
Annotations the server declares
- title: Inspect China data publication rights
- readOnlyHint: true
- destructiveHint: false
- idempotentHint: true
- openWorldHint: false
list_signalsList every published signal Palimpsest exposes across its three applications: name, one-line description and source URL for each.Reads text others writeNo findings
Why it is in this group
- The name "list_signals" says "list": a word for reading.
What it can reach
- untrusted-inputThe description's first sentence says "list" and "url": it reads web pages, which anyone can write.
Description, as the server wrote it List every published signal Palimpsest exposes across its three applications: name, one-line description and source URL for each. Censorship and information control: OONI Great Firewall probes, Censored Planet, IODA outages, circumvention demand, takedown and redaction pressure, and the board's own verdict. China economics: explicit metadata-only rights status for affected observations, pulse, forecast and derivative surfaces; no denied values are returned. AI model evaluation: the tamper-evident, pre-registered eval registry (hash-chained and Merkle-rooted), its claim-by-claim assurance ceiling, evidence-bound Eval Journal, deterministic live findings, and frontier-model refusal drift, alongside the Generative Firewall Index over a named China-focused panel. Takes no arguments. Call this first to discover signal names, then get_signal for one full reading.
Annotations the server declares
- title: List published signals
- 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 outbound.
- Reads private data0
- Reads files, mail, messages, databases, secrets, calendars or repositories.
- Reads text others write2
- 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, 19:24 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
- 2
- 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 2 tools that change things (2 write). Whether a call also works without credentials was not tested: tools are never called.
What this check looks forMatched text
https://api.seiche.info/palimpsest/mcp: tools/list returned 7 tools to an anonymous client, including research_catalog (write), whats_happening (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.
Tools2 changed.~2 changedView diff
- Changed:gfw_readingOnly readsdescription
- Changed:list_signalsOnly readsdescription
Tools1 added, 1 removed, 6 changed.+1 added−1 removed~6 changedView diff
- Added:inspect_economic_publication_rightsOnly reads
- Removed:query_economic_observations
- Changed:get_newsroomOnly readsannotations
- Changed:get_signalOnly readsannotations
- Changed:gfw_readingOnly readsannotations
- Changed:list_signalsOnly readsannotations
- Changed:research_catalogChanges thingsannotations
- Changed:whats_happeningChanges thingsannotations
Install
Pinned to version 1.9.3. Values in angle brackets are yours to fill in.
claude mcp add --transport http palimpsest https://api.seiche.info/palimpsest/mcpclaude mcp add --transport http palimpsest https://api.seiche.info/palimpsest/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__palimpsest__list_signals",
"mcp__palimpsest__get_signal",
"mcp__palimpsest__get_newsroom",
"mcp__palimpsest__inspect_economic_publication_rights",
"mcp__palimpsest__gfw_reading"
],
"ask": [
"mcp__palimpsest__research_catalog",
"mcp__palimpsest__whats_happening"
],
"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 palimpsest. Review it before you use it.
Capability tags behind 2 rules
- mcp__palimpsest__list_signals untrusted-input
- mcp__palimpsest__gfw_reading 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.beepboop2025/palimpsest/verify?hash=a48774851ce07ceb3b46c335bc5a66fdeb357c91812745a5eec7596d883d13e7" | jq -e '.changed == false'Pinned hash a48774851ce0, from Nov 17, 2026, 09:18 UTC.
Watch
Get told when its tools change: a new tool, a new description, a new class.
https://radar.anlyon.com/servers/io.github.beepboop2025/palimpsest/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://radar.anlyon.com/servers/io.github.beepboop2025/palimpsest)Source: MCP Registry · repository. Synced Oct 21, 2026, 18:23 UTC.
Palimpsest — censorship, China economy and model-eval observatory MCP server: questions and answers
Answered from the scan above, not from the publisher's description.
Is the Palimpsest — censorship, China economy and model-eval observatory MCP server safe?
- No scan can promise that. What its definitions show: 7 tools, 2 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 Palimpsest — censorship, China economy and model-eval observatory MCP server delete data?
- No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.
Can the Palimpsest — censorship, China economy and model-eval observatory 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 Palimpsest — censorship, China economy and model-eval observatory MCP server have?
- 7 tools: 2 write, 5 read. research_catalog, whats_happening, list_signals, get_signal, get_newsroom, inspect_economic_publication_rights and gfw_reading.
Does the Palimpsest — censorship, China economy and model-eval observatory MCP server need sign-in or OAuth?
- Its remote endpoint listed its tools to an anonymous client, without sign-in.
Does the Palimpsest — censorship, China economy and model-eval observatory MCP server send data to outside services?
- None of its tools sends, posts or publishes, or takes a recipient, channel or URL.
Has the Palimpsest — censorship, China economy and model-eval observatory MCP server changed its tools?
- Yes, 2 times since Anlyon Radar started tracking it, most recently on Nov 17, 2026. Every change is on this page and in its Atom feed.