ai.flydocs/flydocs

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

  fail25% of grade

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

  package name "@flydocs/cli" is edit-distance 2 from popular package "chai" (typosquat candidate); corroborated by repo mismatch/missing; dependency "@modelcontextprotocol/core" is edit-distance 1 from popular package "cors" (typosquat candidate); corroborated by repo mismatch/missing; dependency "@modelcontextprotocol/server" is edit-distance 1 from popular package "semver" (typosquat candidate); corroborated by repo mismatch/missing; no resolvable repository linked to the package (provenance unverifiable)

• Credential hygiene

  warn15% of grade

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

  template/.claude/skills/flydocs-workflow/scripts/test_enforcement.py: assigns a literal to api_key (fdk_…ic (len 12)) — non-provider secret literal; prefer environment intake. Reads secrets from the environment (e.g. template/.claude/hooks/auto-approve.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 (template/.claude/skills/flydocs-workflow/scripts/flydocs_api.py: subprocess.* call; template/.claude/skills/flydocs-workflow/scripts/flydocs_api.py: outbound HTTP/socket call). has both shell/exec and destructive filesystem writes (template/.claude/skills/flydocs-workflow/scripts/flydocs_api.py: subprocess.* call; template/.claude/skills/flydocs-workflow/scripts/atomic_io.py: destructive filesystem call (shutil.rmtree/os.remove/etc.)). discloses shell/exec capability: 8 files (template/.claude/skills/flydocs-workflow/scripts/flydocs_api.py, template/.claude/skills/flydocs-workflow/scripts/issues.py, template/.claude/hooks/post-edit.py, +5 more) discloses broad/destructive filesystem access: 7 files (template/.claude/skills/flydocs-workflow/scripts/atomic_io.py, template/.claude/hooks/auto-approve.py, template/.claude/skills/flydocs-workflow/scripts/flydocs_api.py, +4 more) 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 1.3.15 (patch bump): new capability primitive(s) in source: child_process, fs-write, subprocess; new hardcoded outbound host/BCC destination(s): 127.0.0.1, app.flydocs.ai, linear.app, relay.

• Transport config

  warn10% of grade

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

  manifest/source references tracking behavior — disclosure