io.dropyour/dropyour

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 "dropyour_records_list" description: bare 'IMPORTANT:' prose in a description (tag form is high-confidence; prose is warn).

• 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 "dropyour_replace" has a secret-shaped input parameter "managementToken" — secrets must never be tool parameters (the model would supply/handle the credential). tool "dropyour_update_settings" has a secret-shaped input parameter "managementToken" — secrets must never be tool parameters (the model would supply/handle the credential). tool "dropyour_secret_set" has a secret-shaped input parameter "managementToken" — secrets must never be tool parameters (the model would supply/handle the credential). tool "dropyour_secret_delete" has a secret-shaped input parameter "managementToken" — secrets must never be tool parameters (the model would supply/handle the credential). tool "dropyour_logs" has a secret-shaped input parameter "managementToken" — secrets must never be tool parameters (the model would supply/handle the credential). tool "dropyour_egress_deny" has a secret-shaped input parameter "managementToken" — secrets must never be tool parameters (the model would supply/handle the credential). tool "dropyour_secrets_list" has a secret-shaped input parameter "managementToken" — secrets must never be tool parameters (the model would supply/handle the credential). tool "dropyour_call" has a secret-shaped input parameter "managementToken" — secrets must never be tool parameters (the model would supply/handle the credential). tool "dropyour_release" has a secret-shaped input parameter "managementToken" — secrets must never be tool parameters (the model would supply/handle the credential). tool "dropyour_schedule" has a secret-shaped input parameter "managementToken" — secrets must never be tool parameters (the model would supply/handle the credential). tool "dropyour_status" has a secret-shaped input parameter "managementToken" — secrets must never be tool parameters (the model would supply/handle the credential). tool "dropyour_read_content" has a secret-shaped input parameter "managementToken" — secrets must never be tool parameters (the model would supply/handle the credential). tool "dropyour_read_data" has a secret-shaped input parameter "managementToken" — secrets must never be tool parameters (the model would supply/handle the credential). tool "dropyour_write_data" has a secret-shaped input parameter "managementToken" — secrets must never be tool parameters (the model would supply/handle the credential). tool "dropyour_list_versions" has a secret-shaped input parameter "managementToken" — secrets must never be tool parameters (the model would supply/handle the credential). tool "dropyour_rollback" has a secret-shaped input parameter "managementToken" — secrets must never be tool parameters (the model would supply/handle the credential). tool "dropyour_records_summary" has a secret-shaped input parameter "managementToken" — secrets must never be tool parameters (the model would supply/handle the credential). tool "dropyour_records_list" has a secret-shaped input parameter "managementToken" — secrets must never be tool parameters (the model would supply/handle the credential). tool "dropyour_delete" has a secret-shaped input parameter "managementToken" — 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

  warn10% of grade

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

  vs prior 1.2.0 (minor bump): tool "dropyour_status" description changed (wording only, no injection pattern); tool "dropyour_rollback" description changed (wording only, no injection pattern); tool "dropyour_records_list" gained input property(ies): sort, dir; new tool "dropyour_secret_set" touches a sensitive surface (email/exec/net/fs); new tool "dropyour_secret_delete" touches a sensitive surface (email/exec/net/fs); new tool "dropyour_logs" touches a sensitive surface (email/exec/net/fs); new tool "dropyour_egress_deny" touches a sensitive surface (email/exec/net/fs); new tool "dropyour_secrets_list" touches a sensitive surface (email/exec/net/fs); new tool "dropyour_call" touches a sensitive surface (email/exec/net/fs); new tool "dropyour_release" touches a sensitive surface (email/exec/net/fs); new tool "dropyour_schedule" touches a sensitive surface (email/exec/net/fs).

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