com.energyaisolution/energyai

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 "get_builder_upgrade_link" description: bare 'IMPORTANT:' prose in a description (tag form is high-confidence; prose is warn). tool "find_local_installers" 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 "run_power_screen" has a secret-shaped input parameter "apiKey" — secrets must never be tool parameters (the model would supply/handle the credential). tool "run_power_screen" has a secret-shaped input parameter "quoteToken" — secrets must never be tool parameters (the model would supply/handle the credential). tool "bootstrap_energy_project" has a secret-shaped input parameter "apiKey" — secrets must never be tool parameters (the model would supply/handle the credential). tool "review_installer_quote" has a secret-shaped input parameter "apiKey" — 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

  fail10% of grade

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

  vs prior 1.0.3 (patch bump): tool "get_quote_link" description changed (wording only, no injection pattern); tool "route_lead" description changed (wording only, no injection pattern); tool "create_builder_key" description changed (wording only, no injection pattern); tool "get_builder_upgrade_link" description changed (wording only, no injection pattern); new tool "get_power_passport_link" touches a sensitive surface (email/exec/net/fs); new tool "get_power_service_quote" touches a sensitive surface (email/exec/net/fs); new tool "run_power_screen" touches a sensitive surface (email/exec/net/fs).

• Transport config

  fail10% of grade

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

  remote MCP endpoint declares NO authentication (unauthenticated remote — nginx-ui/CVE-2026-33032 class)