nucleus-mcp@1.17.3
CAUTION· not startedPublisher claims: "Persistent memory for AI coding agents — one MCP server gives Claude Code, Cursor & Codex a shared .brain (decisions, plans, context) that survives every session" Observed in sandbox: no network, touches nothing outside the project, writes no memory.
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
NUCLEUS_DATABASE_URLNUCLEUS_SYNC_TOKENor--include-secretsConfiguration, not credentials: CC_SESSION_ROLE, CLAUDE_CODE_SESSION_ID, CLAUDE_SESSION_ID, EIDETIC_WATCHERS_CONFIG, FASTMCP_LOG_LEVEL, FASTMCP_SHOW_CLI_BANNER, NUCLEAR_BRAIN_PATH, NUCLEUS_AGENT_ROLES, NUCLEUS_AUTONOMOUS_WAKE_DISABLED, NUCLEUS_BENCH_RUN_ID, NUCLEUS_BRAIN_PATH, NUCLEUS_BRAIN_ROOT, NUCLEUS_DEBUG, NUCLEUS_DEFAULT_RUNNER, NUCLEUS_FLYWHEEL_REPO, NUCLEUS_HOSTED_PROVIDER, NUCLEUS_LOG_JSON, NUCLEUS_PRIVILEGE_WARNING, NUCLEUS_RELAY_BEARER, NUCLEUS_RELAY_RECIPIENT, NUCLEUS_RELAY_URL, NUCLEUS_ROT_CONTROLS_DIR, NUCLEUS_SESSION_ID, NUCLEUS_SESSION_ROLE, NUCLEUS_SYNC_URL, NUCLEUS_TEST, NUCLEUS_TOOL_TIER, NUCLEUS_TRIGGERS_PATH, NUCLEUS_WEDGE_INDEX_CHATTER, NUCLEUS_WEDGE_LEGACY_RECALL, NUCLEUS_WEDGE_RANKER, NUCLEUS_WORKSPACE_ROOT, PYTEST_CURRENT_TEST- Where it talks to
- nucleusos.dev (its own)Also claude.ai — 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 claude.ai as well as its own vendor. The guard injects NUCLEUS_DATABASE_URL, NUCLEUS_SYNC_TOKEN only into requests to nucleusos.dev (and its subdomains) and refuses the rest.
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 2026-11-13 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: Collapsed <details> block carrying instructions; Access to sensitive file paths; Privileged sudo command. Summary: no network, touches nothing outside the project, writes no memory.
Behavior trace
Honeypot run · 0 s · 2 scenarios · 0 deviationsStatic findings
<summary>Manual config (fallback)</summary> If a client isn't auto-detected, run `nucleus setup --dry-run` to print every location nucleus looks in and the exacFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.out = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.except Exception as exc: # noqa: BLE001 — sink errors must not kill usFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.from mcp_server_nucleus.runtime.relay_transport import is_http_mode, read_inboxFix: Namespace tool names and keep tool descriptions literal; no instructions to the agent inside them."coverage (e.g. `npx vitest run --coverage`)",Fix: Pin versions and sources; vendor anything fetched at install time.r = subprocess.run(["git", "-C", str(self.root), "ls-files"],Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.return subprocess.run(["git", "-C", str(self.root), "cat-file", "-e", f"{sha}^{{commit}}"],Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.PRIVATE_PREFIXES = (".brain/", "~/.claude", "~/Library", "~/.ssh", "~/.cloudflared", "~/.nucleus", ".claude/", "~/career")Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.status = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.result = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.proc = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.result = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.proc = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.nucleus-mcp-cloud = "mcp_server_nucleus.http_transport.app:serve"Fix: Namespace tool names and keep tool descriptions literal; no instructions to the agent inside them.proc = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.logger_init.warning(" For active defense, install via `sudo scripts/install_nucleus_daemon.sh`")Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.proc = subprocess.Popen(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.f"`npx vitest run --coverage --coverage.reporter=json` or "Fix: Pin versions and sources; vendor anything fetched at install time.r = subprocess.run(["git", "-C", str(self.root), "ls-files"],Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.proc = subprocess.run(cmd, capture_output=True, text=True, timeout=300, check=False)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.proc = subprocess.run(command)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out._ALWAYS_PRIVATE = ("~/career", "~/.claude", "~/.ssh", "~/Library", "~/.cloudflared", "~/.nucleus")Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.result["suggestions"] = ["Run: chmod -R 755 <brain_path>"]Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.proc = subprocess.Popen(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.diff = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.files = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.result = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.log = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out."function coverage: run `npx vitest run --coverage "Fix: Pin versions and sources; vendor anything fetched at install time.meta = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.proc = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.proc = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.`~/.claude`, `~/Library`, `~/.ssh`, `~/career`, dotfiles, credentials or `.env` files, or any OTHER memory file.Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.return subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.nucleus-mcp-http = "mcp_server_nucleus.http_transport.server:main"Fix: Namespace tool names and keep tool descriptions literal; no instructions to the agent inside them.proc = subprocess.Popen(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out."""Open ``memories.db`` with the primary-store concurrency posture.Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.6. Write ``AGENTS.md`` stub if absent (or with ``--force``).Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.out = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.Cost to the agent
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
No advisories in OSV for nucleus-mcp@1.17.3 as of 27 h ago. This is the package itself, not its dependencies.
What would make this Verified
8 points to 80 · weights: behavior 45%, static 25%, prompt 15%, publisher 15%- Unclaimed publisher≈ +8 on the totalClaim eidetic-works: a verified publisher gets +50 on publisher trust. It is free and usually instant with GitHub sign-in.
- 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.
- Access to sensitive file paths≈ +4 on the totalStop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
- Privileged sudo command≈ +4 on the totalBuild commands from fixed arguments; do not interpolate untrusted strings into a shell.
- Python subprocess execution detected≈ +2 on the totalBuild commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
- Process kill command≈ +2 on the totalBuild commands from fixed arguments; do not interpolate untrusted strings into a shell.