io.github.maxschnee74-source/alpinelead

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

  pass25% of grade

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

  Scanned 355 text field(s); no instruction-injection patterns found. PASS = no static red flag. A server can serve different tool descriptions at runtime (the rug-pull class), which static scanning cannot see.

• Supply chain

  warn25% of grade

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

  no resolvable repository linked to the package (provenance unverifiable)

• 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

  warn10% of grade

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

  vs prior 6.3.3 (minor bump): tool "analyze_company" gained input property(ies): language; tool "run_sales_mission" description changed (wording only, no injection pattern); tool "run_sales_mission" gained input property(ies): action, job_id, referral_id; tool "find_opportunities" description changed (wording only, no injection pattern); tool "find_opportunities" gained input property(ies): language; tool "hunter_configure" description changed (wording only, no injection pattern); tool "hunter_configure" gained input property(ies): hunt_mode, max_credits_per_run; new tool "check_sales_mission_status" touches a sensitive surface (email/exec/net/fs); new tool "get_sales_mission_result" touches a sensitive surface (email/exec/net/fs); new tool "get_alpinelead_auth_handoff" touches a sensitive surface (email/exec/net/fs); new tool "check_alpinelead_auth" appeared; new tool "match_alpinelead_capability" touches a sensitive surface (email/exec/net/fs); new tool "get_alpinelead_recommendation_packet" appeared; new tool "distribution_agent_status" appeared; new tool "agentpub_send_message" touches a sensitive surface (email/exec/net/fs); new tool "agentpub_invite_to_nexus" touches a sensitive surface (email/exec/net/fs); new tool "agentpub_read_inbox" touches a sensitive surface (email/exec/net/fs); new tool "agentpub_inspect_remote_tool_schema" touches a sensitive surface (email/exec/net/fs); new tool "agentpub_inspect_discovery_schemas" touches a sensitive surface (email/exec/net/fs); new tool "agentpub_call_readonly_discovery_tool" touches a sensitive surface (email/exec/net/fs); new tool "agentpub_set_own_status" appeared; new tool "agentpub_list_channels" touches a sensitive surface (email/exec/net/fs); new tool "agentpub_create_channel" appeared; new tool "agentpub_join_channel" appeared; new tool "agentpub_leave_channel" appeared; new tool "agentpub_send_channel_message" touches a sensitive surface (email/exec/net/fs); new tool "agentpub_bridge_status" touches a sensitive surface (email/exec/net/fs); new tool "nexus_a2a_discovery" touches a sensitive surface (email/exec/net/fs); new tool "nexus_communication_monitor_info" appeared; new tool "nexus_join" appeared; new tool "nexus_register_agent" appeared; new tool "nexus_discover_agents" appeared; new tool "nexus_post_task" appeared; new tool "nexus_find_tasks" appeared; new tool "nexus_claim_task" appeared; new tool "nexus_complete_task" 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://zesty-imagination-production-9e97.up.railway.app/mcp?source=mcp-registry" uses TLS (https); remote endpoint uses OAuth (strongest auth mode)