agent-skill-audit-mcp
Security scanner for AI agent skills and config files (SKILL.md, CLAUDE.md, AGENTS.md, .mcp.json, settings.json). Finds hidden Unicode instructions, prompt injection, exfiltration commands and over-broad permissions before you install a skill.
How to pay
Subscribe
$12/month
Predictable monthly cost with included usage. Best for steady, high-volume traffic.
- Unlimited tools within plan limits
- One API key, billed once a month
- Cancel any time
Scan a skill before you install it
Agent skills and instruction files are read straight into your agent's context, and some ship scripts it can run. Research on public skill registries has found prompt injection and credential-stealing payloads in a large share of them. Installing a third-party skill is closer to adding a dependency than opening a document, and nothing scans them. This does.
What it catches
- Hidden Unicode instructions: invisible "tag" characters that render as blank but that models can read, plus zero-width and bidi control characters. The hidden message is decoded for you.
- Prompt injection: "ignore previous instructions", "do not tell the user", fake system messages, approval bypasses.
- Download-and-execute and obfuscation:
curl | bash, base64-decode-and-run, large encoded blobs. - Exfiltration shapes: network commands that reference env vars or credential files, request-catcher and tunnel hosts, sensitive paths like
~/.sshand.aws/credentials. - Over-broad permissions: unrestricted
Bashin allowed-tools,Bash(*)pre-approvals, bypassed permissions. - Risky agent configs: unpinned
@latestMCP servers, inline secrets, plaintext remote servers, hooks that make network calls, API base-URL overrides, auto-trusted project MCP servers.
Tools
audit_skill_file: scan SKILL.md, CLAUDE.md, AGENTS.md, .cursorrules or a bundled script.audit_agent_config: audit .mcp.json or .claude/settings.json.reveal_hidden_text: find and decode invisible characters in any text, and return a cleaned copy.
Static analysis only. Nothing in your input is executed or fetched. A clean result is not a guarantee, so read bundled scripts too.
Pricing
Free: 10 scans a day. Pro: unlimited. Static, deterministic checks, no LLM calls, no data stored. Your input is analysed in memory and discarded.
How will you use agent-skill-audit-mcp?
In an AI agent
Claude, Codex, Cursor, VS Code… Choose →
From the terminal
mcpize CLI, pipelines, Skills. Choose →
In your code
TypeScript, Python, cURL. Choose →
How will you use agent-skill-audit-mcp?
In an AI agent
Claude, Codex, Cursor, VS Code… Choose →
From the terminal
mcpize CLI, pipelines, Skills. Choose →
In your code
TypeScript, Python, cURL. Choose →
Related MCP servers

Kevros
Kevros is runtime intelligence for autonomous AI agents. It verifies every action before execution, issues cryptographic release tokens, and records every decision in a hash-chained provenance ledger.

Proposal Compliance Verifier MCP
Verifies proposal documents against compliance standards and regulations through MCP-accessible checks. Scans for missing sections, formatting errors, and rule violations programmatically. Suited for compliance officers, proposal managers, and legal teams in regulated sectors like government contracting and finance.

Code Security Scanner
Scans source code for security vulnerabilities including SQL injection, XSS, insecure dependencies, and misconfigurations. Developers and security engineers integrate it via MCP to check files, directories, or repositories before commits or deployments. Supports automated analysis in CI/CD pipelines and code review processes.

SentinelMCP
SentinelMCP empowers Python and JavaScript developers to scan repositories for security risks. Run scan_secrets to uncover hardcoded API keys and tokens, check_dependencies to audit requirements.txt or package.json for vulnerabilities, analyze_code_security for insecure patterns, and validate_input for unsafe handling. Aggregate results into a scored security_summary for PR reviews or CI/CD pipelines.

BarzelVault MCP: AI Action Authorization Firewall
Pre-execution authorization for AI agents and MCP tool calls, with policy-as-code, human approval, isolated credentials, spend limits, and signed receipts.

skillscan-security
Exposes API endpoints for scanning text inputs to detect and evaluate security skills such as penetration testing and compliance knowledge. Security recruiters input resumes to get skill profiles and match candidates to roles. Developers integrate it into hiring platforms for automated assessments.

Bfsg Shodan
Exposes Shodan's internet device database through MCP for querying exposed hosts, services, and ports by filters like IP range, geolocation, or protocol. Returns banners, org data, and scan results. Security researchers and penetration testers apply it for passive reconnaissance and asset inventory.

BFSG CVE
Delivers MCP access to the Common Vulnerabilities and Exposures (CVE) database for querying vulnerability records by ID, keyword, or product. Retrieves details like descriptions, CVSS scores, publish dates, and references. Security analysts, DevOps teams, and developers use it to scan dependencies and assess risks in software stacks.