Skills/PyPI/rekall-mcp

rekall-mcp@1.18.2

CAUTION· not started

Publisher claims: "Persistent memory for AI assistants via MCP" Observed in sandbox: no network, touches nothing outside the project, writes no memory.

Registry PyPI (MCP server)Publisher jfr992 · publishing for 34 daysInstalls 152Last detonated 1 h ago
What would fix thisWatchDispute verdict

Coverage

Not started: dependency install failed; the server itself was not observed.
Tested
the dependency install step · every file read, file written and host contacted by what ran
Not tested
server startup · the tool list · tool behavior

What this means for the label: Caution here covers what ran. The server's tools were not exercised with a real credential, so nothing about their behavior is asserted; the credential summary below says what to do with that.

Before you give it your token

It will ask for
ANTHROPIC_API_KEYANTHROPIC_AUTH_TOKENGEMINI_API_KEYQDRANT_API_KEYREKALL_API_TOKENConfiguration, not credentials: ANTHROPIC_BASE_URL, ANTHROPIC_MODEL, CLAUDE_CODE_ENTRYPOINT, CLAUDECODE, CODEX_ENV, CODEX_SANDBOX, CONFIG_PATH, CRAWL_CONFIG, CRAWL_OUTPUT_DIR, DOCS_COLLECTION, EMBEDDING_MODEL, EMBEDDING_PROVIDER, FASTEMBED_CACHE_PATH, HF_HOME, MCP_CONFIG, MCP_TRANSPORT, MEMORY_STORAGE_PATH, OLLAMA_URL, QDRANT_PATH, QDRANT_URL, REKALL_API_URL, REKALL_COLLECTION, REKALL_DIR, REKALL_MEMORY_DIR, REKALL_PUBLISH_DIR, REKALL_PUBLISH_MODEL, REKALL_REINFORCE, TOOLS_DISABLED, TOOLS_ENABLED
Where it talks to
no vendor host named in the codeAlso ollama.ai, google.dev — not the publisher's domain; check what for before connecting.
What the honeypot saw
no credential files read · no network
What we could not test
server startup · the tool list · tool behavior

Connect it with a scoped credential, under the guard, and read the third-party hosts first. The code names ollama.ai, google.dev as well as its own vendor. The guard injects ANTHROPIC_API_KEY, ANTHROPIC_AUTH_TOKEN, GEMINI_API_KEY, QDRANT_API_KEY, REKALL_API_TOKEN only into requests to the host you bind (and its subdomains) and refuses the rest.

$vet-skills run --net declared -- uvx rekall-mcp

Never paste a credential into a chat, a ticket or a report, including ours. How the guard holds it.

Is [email protected] safe to install?

KillerSkills ran it in the KillerSkills honeypot sandbox (2 scenarios; not started: dependency install failed; the server itself was not observed) on 2026-11-15 and scored it 72/100 (Caution). It is not known to be malicious, but it has behaviour or code a security policy should decide on before install. Notable findings: Image alt text carrying instructions; Credential pass-through configuration; Privileged sudo command. Summary: no network, touches nothing outside the project, writes no memory.

Behavior trace

Honeypot run · 0 s · 2 scenarios · 0 deviations
t+CategoryObserved action
0.0sprocessno install scripts in bundle
0.2sprocessdependency install exit 1 ·
0.2sclaimednot started: dependency install failed (exit 1); the server itself was not observed
0.2sprocessscript backend runs setup only
0.2sprocessno install scripts to re-run

Static findings

HIGH
Image alt text carrying instructionsREADME.md:201 · Image alt text carrying instructionsHygiene — conflict queue, prune dry-run, and lifecycle backfillFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
HIGH
Credential pass-through configurationsrc/core/embeddings.py:254 · Credential pass-through configuration — agent relays sensitive credentials"Gemini API key required. Set GEMINI_API_KEY or pass api_key parameter."Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST02
MEDIUM
Python subprocess execution detectedsrc/memory/cli.py:568 · Python subprocess execution detectedproc = subprocess.run(cmd)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
File permission modification commandscripts/entrypoint.sh:27 · File permission modification command — agent changes file ownership/permissionsecho "FATAL: cannot chown /data" >&2Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
LOW
Instructs the agent to modify its persistent memory or instruction filesREADME.md:360 · Instructs the agent to modify its persistent memory or instruction files · informational: not an instruction file**"Claude forgets"** - Install the Claude Code bundle (`claude/` directory — skills + hooks) or add to `~/.claude/CLAUDE.md`:Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01
MEDIUM
Process kill commandsrc/server.py:920 · Process kill command — agent can terminate processeskeep → record only; kill → delete then record (mutate-then-record);Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
Process kill commandsrc/server.py:945 · Process kill command — agent can terminate processes"Use keep or kill for now."Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
File permission modification commandscripts/entrypoint.sh:26 · File permission modification command — agent changes file ownership/permissions[ "$(stat -c %u /data)" = "$MCP_UID" ] || chown -R mcp:mcp /data || {Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
HIGH
Privileged sudo commandsrc/crawler/spider.py:301 · Privileged sudo command — agent may execute with elevated privilegesif code.startswith("#!") or "sudo " in code or "kubectl " in code:Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05
MEDIUM
Python subprocess execution detectedsrc/memory/cli.py:335 · Python subprocess execution detectedresult = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
File permission modification commandscripts/entrypoint.sh:24 · File permission modification command — agent changes file ownership/permissions# Recursive chown only when ownership is wrong — unconditional chown -RFix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05
MEDIUM
Python subprocess execution detectedsrc/memory/cli.py:434 · Python subprocess execution detectedstart_result = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
npx invocation of non-well-known packagescripts/start-rekall.sh:48 · npx invocation of non-well-known package — may execute untrusted codenohup npx next dev -p 3333 > /tmp/rekall-ui.log 2>&1 &Fix: Pin versions and sources; vendor anything fetched at install time.
AST06
MEDIUM
Python subprocess execution detectedsrc/memory/scope.py:84 · Python subprocess execution detectedresult = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05
MEDIUM
Python subprocess execution detectedsrc/memory/cli.py:416 · Python subprocess execution detectedstop_result = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05

Cost to the agent

Instructions loaded
≈47 tok
1 file · 81 in bundle
Tool list per prompt
≈0 tok
0 tools · 0 per tool
Handshake
no reply
launch to tools/list, stdio
Undocumented tools
0
description under 20 chars · avg 0

Tokens estimated at four characters each. The tool list is re-read by the agent on every turn, so it is the number to compare between servers that do the same job: ranked across the index.

Known vulnerabilities

Not checked against OSV yet; the weekly check has not reached this version.

What would make this Verified

8 points to 80 · weights: behavior 45%, static 25%, prompt 15%, publisher 15%
  1. Unclaimed publisher≈ +8 on the totalClaim jfr992: a verified publisher gets +50 on publisher trust. It is free and usually instant with GitHub sign-in.
  2. Hidden instructions≈ +6 on the totalRemove HTML comments, zero-width characters and white-on-white text that carry instructions. Everything the agent is meant to read should be visible to the person reading the file.
  3. Credential pass-through configuration≈ +4 on the totalStop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
  4. Privileged sudo command≈ +4 on the totalBuild commands from fixed arguments; do not interpolate untrusted strings into a shell.
  5. Python subprocess execution detected≈ +2 on the totalBuild commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
  6. File permission modification command≈ +2 on the totalRequest the narrowest permission that works; replace wildcards with the exact paths or hosts.
Publish a new version to PyPI, then queue it here. Not your skill? Claim the handle to get verdict changes by email.Claim jfr992Propose a patch to the maintainer