Memem MCP Server
memem
PyPI
v2.9.7
Published by tt-wang — no publish provenance, so origin is unverified, but the source is public: the repository link below is self-declared yet readable, so you can inspect the code before adopting it.
Persistent, self-evolving memory for Claude Code — local-first, browsable Obsidian vault.
The grade answers one question — how safe is this server for you to adopt — so it is computed in two auditable stages. Nothing below is an opinion or an LLM's guess; every line is a real term the deterministic engine applied, and the same input always yields the same number.
1. Threat score — 100 − 6.3 = 94. What the published surface and source actually contain:
| Points | What was found | Category |
|---|---|---|
| −6.3 | Unsafe deserialization MTC-SRC-007 | permissions |
2. Client adoption risk — 94 − 7 = 87. Three small, subtract-only factors that reflect your risk in adopting it — a clean scan proves less on a powerful, unverified or barely-inspectable package, so the grade says so plainly:
| Points | Adoption-risk factor |
|---|---|
| −6 | capability blast radius (high) — client exposure if the model is manipulated |
| −1 | publisher verification (public source) — no provenance, but the source is public and inspectable |
Capability observations and info notes are shown under Findings but never scored.
Open any row's finding below for the file, line and evidence behind a deduction.
Findings 13
high Shell/command execution in server code (memem/capabilities.py)MTC-SRC-002
In the server's implementation (`memem/capabilities.py:70`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: try: result = subprocess.run( ["claude", "--version"], capture_output=True, text=True,
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server memem/capabilities.py
high Shell/command execution in server code (memem/cli.py)MTC-SRC-002
In the server's implementation (`memem/cli.py:452`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: _CODE) result = subprocess.run( [sys.executable, "-m", "memem.mine_delta", "--session-id", sid, "--t
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server memem/cli.py
high Shell/command execution in server code (memem/dreamer.py)MTC-SRC-002
In the server's implementation (`memem/dreamer.py:432`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: result = subprocess.run( ["claude", "-p", "--model", _HAIKU_MODEL_ALIAS, "--tools", "
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server memem/dreamer.py
high Shell/command execution in server code (memem/migrate_layers.py)MTC-SRC-002
In the server's implementation (`memem/migrate_layers.py:168`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: try: result = subprocess.run( [ "claude", "-p", "-
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server memem/migrate_layers.py
high Shell/command execution in server code (memem/mine_delta.py)MTC-SRC-002
In the server's implementation (`memem/mine_delta.py:584`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: try: result = subprocess.run( [ "claude", "-p", "--model", "haiku"
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server memem/mine_delta.py
high Shell/command execution in server code (memem/mining.py)MTC-SRC-002
In the server's implementation (`memem/mining.py:109`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: try: result = subprocess.run( ["claude", "-p", "--model", "haiku", "--tools", "", "--system-prompt
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server memem/mining.py
high Shell/command execution in server code (memem/playbook.py)MTC-SRC-002
In the server's implementation (`memem/playbook.py:137`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: try: result = subprocess.run( ["claude", "-p", "--model", "haiku", "--tools", "", "--system-prompt
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server memem/playbook.py
high Shell/command execution in server code (memem/profiles.py)MTC-SRC-002
In the server's implementation (`memem/profiles.py:355`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: try: result = subprocess.run( ["claude", "-p", "--model", "haiku", "--tools", "", "--system-prompt
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server memem/profiles.py
medium Unsafe deserialization (memem/obsidian_store.py)MTC-SRC-007
In the server's implementation (`memem/obsidian_store.py:216`): Deserializing untrusted data with these APIs can execute arbitrary code (a well-known RCE gadget class). This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: payload = pickle.load(fh) # noqa: S301 except Exception as exc: # noqa: BLE001 log.i
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server memem/obsidian_store.py
low Reads a sensitive credential path or dumps the environment in packaging/dev tooling (tests/test_security.py)MTC-SRC-006
In a packaging/dev/install script (shipped, but not the server runtime) (`tests/test_security.py:29`): Reading private keys / cloud credentials, or serializing the whole environment, is a sensitive-data source that becomes exfiltration when combined with any egress. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: ory_content("cat ~/.ssh/id_rsa") assert result is not None def test_invisible_unicode_blocked(): result = scan
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server tests/test_security.py
low Shell/command execution in packaging/dev tooling (tests/test_bootstrap.py)MTC-SRC-002
In a packaging/dev/install script (shipped, but not the server runtime) (`tests/test_bootstrap.py:28`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: e(env_extra) return subprocess.run( ["bash", str(BOOTSTRAP), *args], capture_output=True, text=True,
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server tests/test_bootstrap.py
low Shell/command execution in packaging/dev tooling (tests/test_obsidian_store.py)MTC-SRC-002
In a packaging/dev/install script (shipped, but not the server runtime) (`tests/test_obsidian_store.py:330`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: nt.parent) result = subprocess.run( [sys.executable, "-c", "import os, sys; " f"os.environ
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server tests/test_obsidian_store.py
low Dynamic module load from a non-literal in packaging/dev tooling (tests/test_operations.py)MTC-SRC-005
In a packaging/dev/install script (shipped, but not the server runtime) (`tests/test_operations.py:40`): Loading a module chosen at runtime (from a variable) can pull in and run attacker-influenced code paths. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: ed 0 memories on second import (dedup), got {second_count}" ) # --------------------------------------------------
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server tests/test_operations.py
Tools 6
Each tool and what it can reach — statically extracted from the published source.
active_memory_sliceno sensitive capabilitymemory_getno sensitive capabilitymemory_saveno sensitive capabilitymemory_searchno sensitive capabilitymemory_timelineno sensitive capabilitytranscript_searchno sensitive capability
What this scan could not see
- Tools were statically extracted from the published source (6 recovered), not enumerated from a running server. Tool-poisoning, Unicode-smuggling, capability and toxic-flow analysis ran on this inferred surface, but a mis-parsed registration could be missed or mis-attributed, so tool-derived findings are capped below “confirmed”. To grade the real runtime surface, scan the running server: --command "npx -y <package>".
Versions 1
Scan history per published version. The engine is deterministic — the same version always yields the same score, so a changed score means the package itself changed.
| Version | Score | Findings | Engine | Scanned |
|---|---|---|---|---|
v2.9.7 latest |
B 87/100 | 13 | 1.13.0 | 2026-08-25 |
Embed this score
Show this server's live Trust Score in your README, docs or website. The badge is served straight from the registry and updates automatically after every rescan — no API key needed. It links back to this page, so anyone who sees the grade can also read the findings behind it instead of taking a number on faith.
Verify this score yourself
The score above is reproducible: the same package version always yields the same result. Run it locally or over the free API — no account, no LLM, fully deterministic.
npx mcptrustchecker scan memem --online --registry pypiMore in Learning & Documentation
stdio compatibility bridge for the hosted AgentMail MCP server
MCP server for Alfa documentation retrieval
MCP server providing Alpine.js documentation and reference
Search the official Astro framework documentation.
MCP server providing ATAG 2.0 (Authoring Tool Accessibility Guidelines) guidance for authoring tool developers
Aurais Research Reader as an MCP server — paste a paper or article, get a structured read with verbatim-verified quotes + a signed Vorion proof chain (via @vorionsys/aurais-core).