io.github.vassiliylakhonin/agenda-intelligence-md

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

  pass25% of grade

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

  Scanned 14 text field(s); no instruction-injection patterns found. PASS = no static red flag. A server can serve different tool descriptions at runtime (the rug-pull class), which static scanning cannot see.

• 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/test_http_access.py: assigns a literal to token (admi…et (len 12)) — non-provider secret literal; prefer environment intake. Reads secrets from the environment (e.g. src/agenda_intelligence/a2a_adapter.py) — the recommended intake shape.

• Permission scope

  warn15% of grade

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

  has both shell/exec and network egress — capable of download-and-run, though no such path was observed (src/agenda_intelligence/cli.py: subprocess.* call; src/agenda_intelligence/cli.py: outbound HTTP/socket call). discloses shell/exec capability: src/agenda_intelligence/cli.py: subprocess.* call 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 1.8.0 (minor bump): new capability primitive(s) in source: dynamic-eval, subprocess; new hardcoded outbound host/BCC destination(s): 127.0.0.1, agenda-intelligence-a2a., agenda-intelligence-a2a.vassiliy-lakhonin.workers.dev, agenstry.com, agent-output-verification-a2a.example.workers.dev, api.opensanctions.org, app.example.com, cdn.tailwindcss.com, evil.example.com, example.com, example.github.io, example.org, example.test, github.com, json.schemastore.org, pypi.org, schemas.openxmlformats.org, watchman.example.com, www.opensanctions.org.

• Transport config

  fail10% of grade

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

  remote MCP endpoint declares NO authentication (unauthenticated remote — nginx-ui/CVE-2026-33032 class); manifest/source references tracking behavior — disclosure