io.github.firebase/firebase-mcp

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

  unscannable25% of grade

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

  No tool descriptions, server instructions, prompts, or resources were fetched; nothing to scan for injection.

• Supply chain

  warn25% of grade

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

  dependency "tar" is edit-distance 2 from popular package "yarn" (typosquat candidate); dependency "tmp" is edit-distance 2 from popular package "pm2" (typosquat candidate); dependency "mime" is edit-distance 2 from popular package "vite" (typosquat candidate); dependency "open" is edit-distance 2 from popular package "boxen" (typosquat candidate); dependency "yaml" is edit-distance 2 from popular package "yarn" (typosquat candidate); dependency "gaxios" is edit-distance 1 from top-20 package "axios" (typosquat candidate); dependency "@apphosting/build" is edit-distance 2 from top-20 package "uuid" (typosquat candidate)

• Credential hygiene

  fail15% of grade

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

  lib/api.js: contains a hardcoded Google API key (AIza…iQ (len 39)). lib/fetchWebSetup.js: assigns a literal to apiKey (fake…ey (len 12)) — non-provider secret literal; prefer environment intake. lib/emulator/auth/handlers.js: assigns a literal to token (${en…)} (len 45)) — non-provider secret literal; prefer environment intake. Reads secrets from the environment (e.g. lib/emulator/adminSdkConfig.js) — the recommended intake shape.

• Permission scope

  warn15% of grade

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

  references credential paths: lib/init/features/aitools/cursor.js: references a credential path (/mcp.json) has both shell/exec and network egress — capable of download-and-run, though no such path was observed (lib/emulator/commandUtils.js: imports child_process and calls exec/spawn; lib/apiv2.js: outbound HTTP/socket call). has both shell/exec and destructive filesystem writes (lib/emulator/commandUtils.js: imports child_process and calls exec/spawn; lib/commands/appdistribution-testcases-export.js: destructive/writing fs.* call (write/unlink/rm/chmod/rename)). discloses shell/exec capability: 11 files (lib/emulator/commandUtils.js, lib/init/features/dataconnect/create_app.js, lib/emulator/dataconnectEmulator.js, +8 more) discloses dynamic code execution: lib/emulator/functionsEmulatorRuntime.js: uses new Function(…) discloses broad/destructive filesystem access: 23 files (lib/commands/appdistribution-testcases-export.js, lib/commands/apps-sdkconfig.js, lib/management/apps.js, +20 more) Capability DISCLOSURE, not a verdict: static analysis sees the primitive is present and reachable, not whether its use is attacker-controlled.

• 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.firebase/[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

  warn10% of grade

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

  manifest/source references analytics behavior — disclosure