ai.getminds/minds

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 "export_study" param "scheduledRunId" description: cross-tool redirection ('instead of call…').

• Supply chain

  pass25% of grade

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

  repository linkage verified

• Credential hygiene

  pass15% of grade

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

  No hardcoded secrets, secret-shaped tool parameters, or leaked keys found. PASS = no static red flag; static analysis cannot prove the code honors env-based secret handling at runtime.

• 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

  fail10% of grade

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

  vs prior 2.0.1 (patch bump): tool "list_research_methods" description changed (wording only, no injection pattern); tool "list_study_drafts" description changed (wording only, no injection pattern); tool "save_study_draft" description changed (wording only, no injection pattern); tool "save_study_draft" gained input property(ies): audienceIds; new tool "create_study" touches a sensitive surface (email/exec/net/fs); new tool "ask_study" touches a sensitive surface (email/exec/net/fs); new tool "export_study" appeared; new tool "export_mind" appeared; new tool "export_audience" appeared; new tool "export_heatmap" appeared; new tool "list_studies" appeared; new tool "get_study_status" appeared; new tool "list_audiences" appeared; new tool "get_audience_limits" touches a sensitive surface (email/exec/net/fs); new tool "create_audience_from_brief" appeared; new tool "ask_audience" touches a sensitive surface (email/exec/net/fs); new tool "plan_study_questions" touches a sensitive surface (email/exec/net/fs); new tool "run_study_questions" touches a sensitive surface (email/exec/net/fs); new tool "get_study_run" touches a sensitive surface (email/exec/net/fs); new tool "get_study_summary" appeared.

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