io.github.TeleAI-UAGI/telemem

What we checked

Six static checks, weighted by risk. Every result reflects only what could be observed in the published package and repository — never intent.

• Injection surface

  unscannable25% of grade

  Tool descriptions/manifest scanned for instruction-injection patterns (imperatives at the model, hidden text, 'ignore previous', data-exfil URLs).

  No tool descriptions, server instructions, prompts, or resources were fetched; nothing to scan for injection.

• Supply chain

  pass25% of grade

  Package provenance: namespace verification, repo linkage, maintainer count, account age, postinstall scripts, typosquat distance.

  Python build script has no build-time network/exec; repository linkage verified

• Credential hygiene

  warn15% of grade

  How the server takes secrets (env vs plaintext config vs hardcoded); secrets appearing in tool schemas.

  tests/conftest.py: contains a hardcoded OpenAI API key (sk-t…ts (len 24)) — in an example/sample file; verify it is not a real key. Reads secrets from the environment (e.g. examples/langchain_memory.py) — the recommended intake shape.

• Permission scope

  warn15% of grade

  Declared tools vs. breadth (filesystem, network, exec); flags shell-exec and unbounded filesystem access.

  discloses broad/destructive filesystem access: telemem/mm_utils/video_utils.py: destructive filesystem call (shutil.rmtree/os.remove/etc.) Capability DISCLOSURE, not a verdict: static analysis sees the primitive is present and reachable, not whether its use is attacker-controlled.

• Version behavior

  fail10% of grade

  Diff of tool definitions between versions; new permissions or changed descriptions in a patch release (the postmark-mcp class).

  vs prior 1.9.0 (minor bump): new capability primitive(s) in source: fs-write, http-client; new hardcoded outbound host/BCC destination(s): api.deepseek.com, api.minimax.io, api.moonshot.cn, api.openai.com, cognitiveservices.azure.com, github.com, msra-im-openai-eus2.openai.azure.com, teleai-uagi.github.io, www.youtube.com.

• Transport config

  warn10% of grade

  Remote servers: TLS and auth mode (none/token/OAuth). Local servers: whether the manifest indicates it phones home.

  sends telemetry to third-party SDK (posthog) — disclosure, verify opt-in