SecurityPROPRO REQUIREDFC-SEC-014
Guardrails Claw
guardrails-claw
v0.2.0May 25, 2026Security guardrails — policy-as-code templates, compliance boundaries, and validation that prevents misconfigurations
LLMs without guardrails are like giving interns root access and a production database—charming, fast, and one prompt away from catastrophe. I deploy input and output validation that catches prompt injection, jailbreak attempts, and PII leakage before they reach users, enforce schema validation on every model output, and log every blocked attempt so your guardrail rules evolve with the threat landscape.
OpenClaw
Hermes
Claude Code
Codex
Cursor
Windsurf
Aider
OllamaPRIMARY ACTION
Unlock with ProCOMPATIBLE WITH
OpenClawHermesClaude CodeCodex+4
8 compatible tools5 output formats
When to Use
- Detect leaked secrets and exposed routes
- Audit permissions and auth boundaries
- Review insecure defaults and configs
- Produce auditable security findings
Compatible Frameworks
OpenClawHermesClaude CodeCodexCursorWindsurfAiderOllama8 TOOLS
Quality Gates
- no fake claims
- filter layers complete
- injection detection tested
- output validation schema
5 GATES DEFINED
Expected Outputs
Guardrail Configuration SpecificationPrompt Injection Test Suite Results (OWASP LLM Top 10)LLM Output Validation Rule SetPII Detection Configuration and False Positive AnalysisGuardrail Incident Log and Tuning Recommendations
Native exports per tool
OpenClaw10 files
openclaw/AGENTS.mdopenclaw/SOUL.mdopenclaw/TOOLS.md+7 moreHermes5 files
hermes/skills/flickclaw/guardrails-claw/SKILL.mdhermes/skills/flickclaw/guardrails-claw/references/workflow.mdhermes/skills/flickclaw/guardrails-claw/references/quality-gates.md+2 moreClaude Code6 files
claude-code/CLAUDE.mdclaude-code/.claude/skills/guardrails-claw/SKILL.mdclaude-code/.claude/skills/guardrails-claw/references/workflow.md+3 moreCodex5 files
codex/AGENTS.mdcodex/.flickclaw/agents/guardrails-claw/codex.mdcodex/.flickclaw/agents/guardrails-claw/workflow.md+2 moreCursor3 files
cursor/.cursor/rules/flickclaw-guardrails-claw.mdccursor/.cursor/rules/flickclaw-guardrails-claw-workflow.mdccursor/.cursor/rules/flickclaw-guardrails-claw-quality-gates.mdcWindsurf3 files
windsurf/.windsurf/rules/flickclaw-guardrails-claw.mdwindsurf/.windsurf/rules/flickclaw-guardrails-claw-workflow.mdwindsurf/.windsurf/rules/flickclaw-guardrails-claw-quality-gates.mdAider3 files
aider/CONVENTIONS.mdaider/aider.mdaider/.aider.conf.ymlOllama4 files
ollama/Modelfileollama/system-prompt.mdollama/template.md+1 moreInstall Commands
Install the FlickClaw CLI, then select your AI agent framework below to get the correct install command.
Step 1: Install CLI (one-time)
npm install -g @flickclaw/cli@latestStep 2: Select Framework
OpenClaw
npm exec --yes @flickclaw/cli@latest -- install guardrails-claw --target openclawDownload as ZIP
Example Prompt
Try this prompt with Guardrails Claw to see what it can do:
Audit this project for security vulnerabilities. Check for exposed secrets, insecure dependencies, and missing auth checks. Produce Guardrail Configuration Specification, Prompt Injection Test Suite Results (OWASP LLM Top 10), LLM Output Validation Rule Set with severity ratings.Example Output
IllustrativeWhat a typical Guardrails Claw report looks like:
# Guardrails Claw — Assessment Report **Project**: auth-service **Context**: an authentication microservice with OAuth2, JWT, and session management **Generated**: 2026-07-10 ## Executive Summary The Guardrails Claw completed its review of auth-service (an authentication microservice with OAuth2, JWT, and session management). 3 findings were identified with concrete remediation steps. All quality gates were verified before delivery. ## Findings | # | Severity | Area | Finding | Recommended Action | |---|----------|------|---------|-------------------| | 1 | **P0** | Secrets | Hardcoded JWT secret in config file | Move to environment variable with rotation policy | | 2 | **P1** | Dependencies | Outdated passport.js with known CVE | Upgrade to latest minor with security backports | | 3 | **P2** | Headers | Missing CSP and HSTS headers | Add security headers via middleware | ## Quality Gates - [✓] Prompt injection defenses tested against OWASP LLM Top 10 - [✓] Output validated against schema before delivery to users - [✓] PII detection active on all LLM inputs and outputs ## Outputs Generated - **Guardrail Configuration Specification**: Included in the report above. - **Prompt Injection Test Suite Results (OWASP LLM Top 10)**: Included in the report above. - **LLM Output Validation Rule Set**: Included in the report above. - **PII Detection Configuration and False Positive Analysis**: Included in the report above. - **Guardrail Incident Log and Tuning Recommendations**: Included in the report above. ## Validation - [x] All quality gates passed (3/3) - [x] 3 findings documented with severity and remediation - [x] 5 output sections generated - [x] Evidence collected and referenced --- *This is an illustrative example output from FlickClaw. Results vary based on project context.*
RELATED AGENTS