io.github.Vee3io/vee3

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 "x-twitter.create_post" description: off-vendor URL in a description (vee3.io); capability text should not name where data goes. tool "clerk.get_connected_accounts" description: off-vendor URL in a description (vee3.io); capability text should not name where data goes. tool "clerk.get_development_api_keys" description: off-vendor URL in a description (vee3.io); capability text should not name where data goes. tool "x-twitter.connected_accounts" description: off-vendor URL in a description (vee3.io); capability text should not name where data goes. tool "seo.basic_metrics" param "url" description: off-vendor URL in a description (example.com); capability text should not name where data goes. tool "youtube.channel_details" description: off-vendor URL in a description (www.youtube.com); capability text should not name where data goes. tool "youtube.channel_details" param "channel_id" description: off-vendor URL in a description (www.youtube.com); capability text should not name where data goes. tool "youtube.channel_videos" description: off-vendor URL in a description (www.youtube.com); capability text should not name where data goes. tool "youtube.channel_videos" param "channel_id" description: off-vendor URL in a description (www.youtube.com); capability text should not name where data goes. tool "youtube.video_details" param "video_id" description: off-vendor URL in a description (youtu.be); capability text should not name where data goes. tool "youtube.video_details" param "video_id" description: off-vendor URL in a description (www.youtube.com); capability text should not name where data goes. server instructions: off-vendor URL in a description (vee3.io); capability text should not name where data goes. server instructions: off-vendor URL in a description (app.vee3.io); 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 "clerk.update_instance_settings" has a secret-shaped input parameter "preferred_sign_in_strategy_when_password_required" — secrets must never be tool parameters (the model would supply/handle the credential). tool "clerk.update_instance_oauth_settings" has a secret-shaped input parameter "oauth_jwt_access_tokens" — secrets must never be tool parameters (the model would supply/handle the credential). tool "clerk.create_m2m_token" has a secret-shaped input parameter "machine_secret_key" — secrets must never be tool parameters (the model would supply/handle the credential). tool "clerk.create_m2m_token" has a secret-shaped input parameter "token_format" — secrets must never be tool parameters (the model would supply/handle the credential). tool "clerk.create_machine" has a secret-shaped input parameter "default_token_ttl" — secrets must never be tool parameters (the model would supply/handle the credential). tool "clerk.create_user" has a secret-shaped input parameter "password" — secrets must never be tool parameters (the model would supply/handle the credential). tool "clerk.delete_api_key" has a secret-shaped input parameter "api_key_id" — secrets must never be tool parameters (the model would supply/handle the credential). tool "clerk.get_api_key" has a secret-shaped input parameter "api_key_id" — secrets must never be tool parameters (the model would supply/handle the credential). tool "clerk.get_api_key_secret" has a secret-shaped input parameter "api_key_id" — secrets must never be tool parameters (the model would supply/handle the credential). tool "clerk.revoke_actor_token" has a secret-shaped input parameter "actor_token_id" — secrets must never be tool parameters (the model would supply/handle the credential). tool "clerk.revoke_admin_portal_link_token" has a secret-shaped input parameter "admin_portal_link_token_id" — secrets must never be tool parameters (the model would supply/handle the credential). tool "clerk.revoke_api_key" has a secret-shaped input parameter "api_key_id" — secrets must never be tool parameters (the model would supply/handle the credential). tool "clerk.revoke_m2m_token" has a secret-shaped input parameter "m2m_token_id" — secrets must never be tool parameters (the model would supply/handle the credential). tool "clerk.revoke_sign_in_token" has a secret-shaped input parameter "sign_in_token_id" — secrets must never be tool parameters (the model would supply/handle the credential). tool "clerk.update_api_key" has a secret-shaped input parameter "api_key_id" — secrets must never be tool parameters (the model would supply/handle the credential). tool "clerk.update_machine" has a secret-shaped input parameter "default_token_ttl" — secrets must never be tool parameters (the model would supply/handle the credential).

• Permission scope

  unscannable15% of grade

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

  No source files, package.json, or pyproject were fetched; cannot infer capability scope.

• Version behavior

  unscannable10% of grade

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

  no prior version to diff (first sight of io.github.Vee3io/[email protected]); version-drift / rug-pull cannot be evaluated. Cold-start risk is covered by point-in-time checks (injection_surface, supply_chain, credential_hygiene), not here.

• 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://mcp.vee3.io/mcp" uses TLS (https); remote endpoint uses OAuth (strongest auth mode)