io.github.justjlee/omnarai-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

  warn25% of grade

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

  server instructions: off-vendor URL in a description (omnarai.vercel.app); capability text should not name where data goes.

• Supply chain

  pass25% of grade

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

  repository linkage verified

• Credential hygiene

  pass15% of grade

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

  Reads secrets from the environment (e.g. index.js) — the recommended intake shape. PASS = no static red flag; static analysis cannot prove the code honors env-based secret handling at runtime.

• Permission scope

  fail15% of grade

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

  HIGH: shell/exec capability AND network egress together — download-and-execute risk (test/tool-parity.test.js: imports child_process and calls exec/spawn; index.js: outbound HTTP/socket call). HIGH: shell/exec capability AND destructive filesystem writes together (test/tool-parity.test.js: imports child_process and calls exec/spawn; scripts/check-tool-parity.js: path traversal ('../') combined with file access). discloses shell/exec capability: test/tool-parity.test.js: imports child_process and calls exec/spawn discloses broad/destructive filesystem access: scripts/check-tool-parity.js: path traversal ('../') combined with file access discloses broad/destructive filesystem access: test/decision-handoff.test.js: path traversal ('../') combined with file access discloses broad/destructive filesystem access: test/decision-state.test.js: path traversal ('../') combined with file access discloses broad/destructive filesystem access: test/decision-store.test.js: path traversal ('../') combined with file access discloses broad/destructive filesystem access: test/info-format.test.js: path traversal ('../') combined with file access discloses broad/destructive filesystem access: test/inquiry-brief.test.js: path traversal ('../') combined with file access discloses broad/destructive filesystem access: test/tool-parity.test.js: path traversal ('../') combined with file access discloses broad/destructive filesystem access: scripts/publish.sh: destructive/writing fs.* call (write/unlink/rm/chmod/rename) 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 1.6.1 (minor bump): new tool "omnarai_context" appeared; new tool "omnarai_divergence" touches a sensitive surface (email/exec/net/fs); new tool "omnarai_inquiry_brief" appeared; new tool "omnarai_query" appeared; new tool "omnarai_trace" appeared; new tool "omnarai_job" appeared; new tool "omnarai_council" touches a sensitive surface (email/exec/net/fs); new tool "omnarai_info" appeared; new capability primitive(s) in source: child_process, fs-write, http-client; new hardcoded outbound host/BCC destination(s): engine.test, huggingface.co, omnarai.vercel.app, registry.modelcontextprotocol.io, www.npmjs.com.

• 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); manifest/source references telemetry behavior — disclosure