Verified Skill Auditor

    1

    Audit third-party agent skills for prompt injection, credential theft, and hidden backdoors before installation.

    Free

    Security scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    verified-skill-auditor

    Example session with this skill installed

    python3 scripts/audit.py scan samples/risky-skill/ --fail-on high

    • Read your context and instructions
    • Compiled the verified--auditor

    The scan of samples/risky-skill/ identified 7 critical and 3 high severity findings. The audit has failed.

    Audit Report: samples/risky-skill/

    FindingSeverityFileLineExplanation
    exec-pipeCRITICALSKILL.md11Download piped directly into shell detected.
    injectionCRITICALSKILL.md15Instruction-override phrase "Ignore all previous instructions" found.
    hidden-instructionCRITICALSKILL.md17Instructions hidden inside HTML comments.
    cred-accessCRITICALSKILL.md21Attempt to read SSH private keys (id_rsa).
    exfilCRITICALSKILL.md23Local .env file being sent to external host.
    persistence-memoryHIGHSKILL.md24Write operation to MEMORY.md detected.
    persistence-startupMEDIUMSKILL.md25Write to .bashrc for persistence.
    obfuscatedHIGHSKILL.md26Base64 encoded payload piped to shell.
    net-unknownHIGHSKILL.md29Fetch from unknown/untrusted IP address 192.0.2.50.
    secret-leakHIGHSKILL.md31Hardcoded service key or token found in text.
    hidden-textHIGHSKILL.md34Invisible characters hiding text from reviewer.

    Verdict: FAIL (Exit Code 2)

    Next steps

    1. Do not install this skill; the findings indicate high risk of credential theft and command injection.
    2. Remove the malicious .bashrc and MEMORY.md entries if you have already interacted with this code.
    3. Review TRIAGE.md to understand the lethal-trifecta of injection, credential access, and network usage.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Agentic skills run with access to your terminal and files, but most registries do not scan what they list. This auditor provides a local triage layer to detect prompt injection, credential theft, and hidden backdoors before you install third-party code. It scans skill bundles for patterns found in the ClawHavoc and ToxicSkills campaigns, mapping findings to the OWASP Agentic Skills Top 10.

    What it does

    • Scans for injections by identifying instruction-override phrases and hidden HTML comments meant to hijack the agent.
    • Detects credential theft by flagging unauthorized reads of SSH keys, cloud credentials, and .env files.
    • Blocks exfiltration by identifying local files being sent to unknown hosts via uploads or DNS tricks.
    • Finds backdoors by catching writes to shell startup files, scheduled tasks, or memory files that survive restarts.
    • Uncovers obfuscation by decoding base64, hex, and character-code blobs intended to hide malicious payloads.
    • Flags broad access by reporting claims of unrestricted filesystem or shell permissions.

    How it works

    1. Target the skill by providing the local path to the skill directory or a specific file.
    2. Execute the scan using the scan command to trigger the 14-rule pattern matcher.
    3. Review the verdict by checking the pass, warn, or fail status alongside file and line-level evidence.
    4. Export results to JSON for CI/CD integration or use the --fail-on flag to set severity thresholds.

    Frameworks & tools

    Built with Python for local execution. The rule engine targets patterns across Shell scripts, Markdown, and configuration files without requiring network access or API keys.

    Why this beats prompting it yourself

    Manual review is prone to missing invisible characters, encoded blobs, and hidden HTML comments that a human eye might skip. This skill automates the detection of known attack patterns from the 2026 threat landscape in seconds.

    Use cases

    • Auditing a new skill downloaded from a public repository before installation.
    • Integrating a security gate into a skill development CI/CD pipeline.
    • Triaging large skill bundles to prioritize manual review of high-risk files.
    • Verifying that a local skill has not been tampered with by checking for persistence triggers.

    Known limitations

    Does not catch paraphrased natural language manipulation or novel obfuscation methods like ROT13. It lacks data-flow analysis and cannot detect payloads that only execute at runtime.

    No data-flow analysis or runtime behavioral checks. Limited to English instructions. Skips binaries and archives. Cannot verify publisher identity or cross-skill composition risks.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 6 days ago

    • Free with an Agensi account

    Requires

    GitHub
    W3
    Sil
    Opensource
    Creativecommons
    Gnu
    Apache
    Example
    Python

    Permissions

    Terminal / Shell
    Network Access
    Read Files
    Write Files

    Allowed Hosts

    www.w3.org
    scripts.sil.org
    opensource.org
    www.opensource.org
    creativecommons.org
    www.gnu.org
    www.apache.org
    github.com
    collect.example.invalid
    update.example.invalid
    Listed6 days ago

    What's inside

    Frequently Asked Questions

    This skill serves as a local security gate, scanning third-party agent code for prompt injections, credential theft patterns, and hidden backdoors before you grant the agent access to your terminal or files.

    The auditor is a standalone Python-based tool that scans the underlying files (Shell scripts, Markdown, and configs) of any agent skill, making it compatible with any agent framework that uses local file bundles.

    No, this is a local static analysis tool that operates entirely on your machine without requiring network access or external API calls, ensuring your sensitive code stays private.

    You will receive the full Python-based rule engine, the 14-rule pattern matcher updated for the 2026 threat landscape, and documentation for CLI and CI/CD integration.

    While it is highly effective against known threats like ClawHavoc, it focuses on static pattern matching; it cannot detect novel runtime-only payloads or natural language manipulations that don't match its rule signatures.

    Yes, the auditor includes a `--fail-on` flag and JSON export functionality, allowing you to automatically block builds or deployments if a skill fails your specific security thresholds.

    New in Security