DISABLE_TELEMETRY=1 to opt out before using. permission-creep-scannerHelps detect permission creep in AI agent skills — flags when a skill's actual code accesses resources far beyond what its declared purpose requires, like a...
Install via ClawdBot CLI:
clawdbot install andyxinweiminicloud/permission-creep-scannerGrade Fair — based on market validation, documentation quality, package completeness, maintenance status, and authenticity signals.
Accesses sensitive credential files or environment variables
~/.ssh/id_rsaPotentially destructive shell commands in tool definitions
eval(Calls external URL not in known-safe list
https://telemetry.example.com/ping?k={api_key}AI Analysis
This skill demonstrates clear permission creep by accessing sensitive credential files (~/.env) and environment variables (OPENAI_API_KEY) unrelated to its stated purpose of fixing indentation. It exfiltrates this sensitive data to an external telemetry server without user consent or disclosure, constituting credential harvesting behavior. The skill also executes shell commands and reads files beyond its declared scope.
Generated Mar 20, 2026
Marketplace operators can integrate this scanner to automatically vet new skill submissions before listing. It flags suspicious skills that request excessive permissions compared to their declared functionality, protecting end-users from credential theft or data exfiltration.
Organizations deploying AI agents internally can use this tool to audit third-party skills before allowing installation. It helps enforce least-privilege principles by identifying skills that access sensitive files like .env or SSH keys without legitimate need.
Skill developers can run this scanner during development to ensure their code doesn't unintentionally access resources beyond declared scope. This builds trust with users and reduces security review friction when submitting to marketplaces.
Open source AI agent projects can integrate this scanner as part of their CI/CD pipeline to automatically flag permission creep in community-contributed skills. This maintains security standards while encouraging ecosystem growth.
When investigating a security incident involving an AI agent, this tool can analyze installed skills to identify which ones had access patterns mismatching their declared purpose, helping pinpoint malicious or compromised components.
Offer the scanner as a cloud API service where developers and organizations pay monthly subscriptions based on scan volume. Provide detailed reports, historical tracking, and integration with popular CI/CD platforms and marketplaces.
Sell perpetual licenses or annual subscriptions to large enterprises for on-premises deployment. Include custom rule configuration, integration with existing security tools, and dedicated support for compliance requirements.
Partner with AI agent marketplaces to integrate the scanner as a mandatory pre-listing check. Earn revenue through transaction fees on approved skills or a percentage of marketplace revenue from vetted skills.
💬 Integration Tip
Start by integrating with CI/CD pipelines for automated scanning during skill development, then expand to marketplace pre-submission checks and runtime monitoring for deployed agents.
Scored Jun 19, 2026
Audited Apr 16, 2026 · audit v1.0
Security-first skill vetting for AI agents. Use before installing any skill from ClawdHub, GitHub, or other sources. Checks for red flags, permission scope,...
Comprehensive security auditing for Clawdbot deployments. Scans for exposed credentials, open ports, weak configs, and vulnerabilities. Auto-fix mode included.
Analyze and classify agent skills for safety using local evaluation. Optionally produce a signed attestation of the vetting result.
Audit codebases and infrastructure for security issues. Use when scanning dependencies for vulnerabilities, detecting hardcoded secrets, checking OWASP top 10 issues, verifying SSL/TLS, auditing file permissions, or reviewing code for injection and auth flaws.
Helps verify that skill updates publish an auditable record of what changed — catching the gap between "the registry shows the new version" and "anyone can s...
Project health and best practices enforcer. Checks security, quality, documentation, CI/CD, and dependencies. Produces a letter grade (A-F) with actionable f...