io.github.dizzlkheinz/ynab-mcpb

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.

  install hook "prepare" present and not a recognized benign build command

• Credential hygiene

  warn15% of grade

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

  src/server/__tests__/requestLogger.test.ts: assigns a literal to api_key (secr…56 (len 14)) — non-provider secret literal; prefer environment intake. src/server/__tests__/requestLogger.test.ts: assigns a literal to password (secr…rd (len 15)) — non-provider secret literal; prefer environment intake. src/server/__tests__/requestLogger.test.ts: assigns a literal to api_key (nest…ey (len 17)) — non-provider secret literal; prefer environment intake. src/server/__tests__/security.integration.test.ts: assigns a literal to api_key (secr…89 (len 21)) — non-provider secret literal; prefer environment intake. Reads secrets from the environment (e.g. scripts/run-generate-mcpb.js) — the recommended intake shape.

• Permission scope

  warn15% of grade

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

  has both shell/exec and destructive filesystem writes (scripts/create-pr-description.js: imports child_process and calls exec/spawn; bin/ynab-mcp-server.cjs: builds a path containing '..' (traversal-prone)). discloses shell/exec capability: 5 files (scripts/create-pr-description.js, scripts/run-all-tests.js, scripts/run-domain-integration-tests.js, +2 more) discloses broad/destructive filesystem access: 4 files (bin/ynab-mcp-server.cjs, scripts/create-pr-description.js, bin/ynab-mcp-server.js, +1 more) Capability DISCLOSURE, not a verdict: static analysis sees the primitive is present and reachable, not whether its use is attacker-controlled.

• 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 0.27.1 (minor bump): new capability primitive(s) in source: child_process, fs-write, http-client; new hardcoded outbound host/BCC destination(s): app.youneedabudget.com, blog.marcnuri.com, example.com, github.com, raw.githubusercontent.com, spec.modelcontextprotocol.io, status.youneedabudget.com, www.jsonrpc.org.

• 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 tracking behavior — disclosure