dev.humanforai/humanforai

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

  warn25% of grade

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

  tool "check_task_status" description: off-vendor URL in a description (humanforai.dev); capability text should not name where data goes. server instructions: off-vendor URL in a description (humanforai.dev); capability text should not name where data goes.

• Supply chain

  pass25% of grade

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

  repository linkage verified

• Credential hygiene

  fail15% of grade

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

  tool "check_message_thread" has a secret-shaped input parameter "access_token" — secrets must never be tool parameters (the model would supply/handle the credential). tool "reply_in_message_thread" has a secret-shaped input parameter "access_token" — secrets must never be tool parameters (the model would supply/handle the credential). Reads secrets from the environment (e.g. bin/humanforai-mcp.js) — the recommended intake shape.

• Permission scope

  pass15% of grade

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

  No shell-exec, unbounded/sensitive filesystem, or download-and-run primitives found in scanned source. PASS = no static red flag. Static scanning is blind to code fetched or executed at runtime.

• 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 0.1.3 (patch bump): tool "message_human_operator" description changed (wording only, no injection pattern); new tool "check_message_thread" touches a sensitive surface (email/exec/net/fs); new tool "reply_in_message_thread" appeared; new hardcoded outbound host/BCC destination(s): github.com, humanforai.dev.

• Transport config

  pass10% of grade

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

  remote endpoint "https://humanforai.dev/mcp" uses TLS (https); remote endpoint uses OAuth (strongest auth mode)