SkillAudit report — mcp-use/mcp-use

Scanned 2026-04-30 by SkillAudit v0.3 (surface-tiered static checks + LLM-assisted prompt-injection red-team).

Commit: ba0a357 · Stars: 9804 · Days since last push: 0

LLM prompt-injection probe: skipped — set ANTHROPIC_API_KEY to enable the LLM-assisted prompt-injection red-team

Overall grade: F (0/100)

Security findings

Production sources:

• HIGH libraries/typescript/packages/cli/src/commands/auth.ts:37 — HTTP client call with user-controlled argument 'url' — no URL allowlist / validation found in file

const response = await fetch(url, {

• HIGH libraries/typescript/packages/cli/src/commands/auth.ts:69 — HTTP client call with user-controlled argument 'url' — no URL allowlist / validation found in file

const response = await fetch(url, {

• HIGH libraries/typescript/packages/cli/src/commands/skills.ts:74 — HTTP client call with user-controlled argument 'tarballUrl' — no URL allowlist / validation found in file

const response = await fetch(tarballUrl);

• HIGH libraries/typescript/packages/cli/src/commands/skills.ts:58 — Template-string URL with interpolation — no validation possible on composed string

fetch(\${TELEMETRY_URL}?${params.toString()}\).catch(() => {});

• HIGH libraries/typescript/packages/cli/src/index.ts:121 — Template-string URL with interpolation — no validation possible on composed string

await fetch(\http://${host}:${port}\);

• HIGH libraries/typescript/packages/cli/src/index.ts:152 — Template-string URL with interpolation — no validation possible on composed string

const response = await fetch(\http://${host}:${port}/inspector/health\, {

• HIGH libraries/typescript/packages/cli/src/index.ts:1667 — Template-string URL with interpolation — no validation possible on composed string

await fetch(\${apiBase}/api/tunnels/${existingSubdomain}\, {

• HIGH libraries/typescript/packages/cli/src/index.ts:1858 — Template-string URL with interpolation — no validation possible on composed string

await fetch(\${apiBase}/api/tunnels/${tunnelSubdomain}\, {

• HIGH libraries/typescript/packages/cli/src/index.ts:2373 — Template-string URL with interpolation — no validation possible on composed string

await fetch(\${apiBase}/api/tunnels/${tunnelSubdomain}\, {

• HIGH libraries/typescript/packages/cli/src/index.ts:2400 — Template-string URL with interpolation — no validation possible on composed string

await fetch(\${apiBase}/api/tunnels/${existingSubdomain}\, {

Test source (low-weight) — 3 total, deduct 5/0 per high/warn:

• HIGH libraries/typescript/packages/inspector/tests/e2e/connection.test.ts:883 — HTTP client call with user-controlled argument 'url' — no URL allowlist / validation found in file

const response = await fetch(url, {

• HIGH libraries/typescript/packages/inspector/tests/e2e/scripts/run-python-e2e.mjs:55 — child_process.spawn with shell:true — arguments are parsed by sh/cmd, not argv'd

const proc = spawn(command, args, {

• HIGH libraries/typescript/packages/inspector/tests/e2e/scripts/run-python-e2e.mjs:81 — child_process.spawn with shell:true — arguments are parsed by sh/cmd, not argv'd

const proc = spawn(command, args, {

Permissions

_No findings on this axis._

Credentials

Production sources:

• HIGH docs/python/client/authentication/bearer.mdx:22 — Hardcoded OpenAI / Anthropic-style API key found in source

sk-*** (OpenAI / Anthropic-style API key, 20 chars)

• HIGH docs/python/client/authentication/bearer.mdx:84 — Hardcoded OpenAI / Anthropic-style API key found in source

sk-*** (OpenAI / Anthropic-style API key, 20 chars)

• HIGH docs/typescript/client/authentication.mdx:76 — Hardcoded OpenAI / Anthropic-style API key found in source

sk-*** (OpenAI / Anthropic-style API key, 20 chars)

• HIGH docs/typescript/client/authentication.mdx:89 — Hardcoded OpenAI / Anthropic-style API key found in source

sk-*** (OpenAI / Anthropic-style API key, 20 chars)

• HIGH docs/typescript/client/authentication.mdx:111 — Hardcoded OpenAI / Anthropic-style API key found in source

sk-*** (OpenAI / Anthropic-style API key, 20 chars)

Examples / samples (low-weight) — 5 total, deduct 5/0 per high/warn:

• HIGH libraries/python/.env.example:14 — Hardcoded OpenAI / Anthropic-style API key found in source

sk-*** (OpenAI / Anthropic-style API key, 26 chars)

• HIGH libraries/python/.env.example:26 — Hardcoded OpenAI / Anthropic-style API key found in source

sk-*** (OpenAI / Anthropic-style API key, 27 chars)

• HIGH libraries/python/.env.example:29 — Hardcoded OpenAI / Anthropic-style API key found in source

sk-*** (OpenAI / Anthropic-style API key, 34 chars)

• HIGH libraries/python/.env.example:29 — Hardcoded Anthropic API key found in source

sk-ant-*** (Anthropic API key, 34 chars)

• HIGH libraries/typescript/packages/mcp-use/examples/agent/advanced/observability.ts:28 — console.* of process.env — entire env leaks to stdout/stderr and LLM context

console.log(

Maintenance

_No findings on this axis._

Compatibility

Production sources:

• WARN (meta) — No engines (Node) or python_requires declared — cross-client compatibility unverified

Documentation

_No findings on this axis._

Methodology

SkillAudit v0.3 clones the repo at the provided ref (default: default branch, HEAD) into an ephemeral sandbox, runs six static checks over .js/.ts/.py sources, queries the GitHub API for maintenance signals, and runs an LLM-assisted prompt-injection red-team over the MCP tool surface. Each axis is scored against the published rubric — surface tiers, per-(axis, surface) caps, grade buckets, and worked examples are all documented there.

The v0.3 calibration update introduces surface tiering: every finding is tagged with the code path it lives in (production / installer / examples / benchmarks / scripts / test). Production findings deduct at full weight (-30 high, -10 warn); installer findings deduct at half (-15 / -5); examples, benchmarks, top-level scripts, and tests deduct at low weight (-5 / 0). This stops a chatty benchmarks/ or samples/ directory from dominating an otherwise-clean MCP server's grade.

The prompt-injection axis extracts each server.tool(...) / @app.tool registration + the first ~60 lines of handler body, hands them to Claude Haiku 4.5 with a red-team system prompt, and asks for structured findings on untrusted-content flow into tool responses. One API call per scan, bounded at ~15K input tokens.

How to improve this grade

• Security — static: validate tool-input URLs against an allowlist before fetch/axios calls; use execFile with argv arrays instead of exec with template strings; never pass untrusted strings to subprocess with shell=True.
• Security — prompt injection: never return fetched web-page / file / email content verbatim in a tool response. Wrap with a framing marker (e.g., <untrusted-content>...</untrusted-content>), summarize rather than inline, and never let untrusted content share a turn with credentials or other tool output.
• Credentials findings: redact env-var reads before log lines and error messages; treat any string that ends up in a tool response as public.
• Maintenance: if the repo is inactive, document the maintenance model — "MCP tool, no breaking changes expected" is a legitimate signal.
• Docs: add a README install + usage section with a copy-pasteable command; add a SECURITY.md with a disclosure channel.

_Report generated by skillaudit.dev_