agentleFS
zyphr-labszyphr-labs/bounty-harness/.claude/skills/agent-safety/SKILL.md

agent-safety

Skill1 starsChanged 3 months agoMITView on GitHub

3 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

1 rule. 0 are common, 0 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 3 edited versions we read keep these.

  • 0

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Before AI triage: sanitize-corpus on any target content.

    line 14

“Common” is measured: kept by at least 6 in 10 of the 3 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Agent Safety

Overview

Protect the AI-assisted bug bounty toolkit from prompt injection in target-controlled content. Sanitize untrusted HTML, JS, markdown, API responses, and tool outputs before they reach the LLM reasoning layer.

Quick Reference

  • Skill: agent-safety
  • Version: 1.0.0
  • Bounded Context: AgentSafetyContext
  • Required tools: python3, jq
  • Risk tier: passive (defensive layer)

Workflow Selection

  • Before AI triage: sanitize-corpus on any target content.
  • During scanning: scan-untrusted-content to detect injection patterns.
  • Review: classify-injection-risk to score content danger.
  • Audit: decision-log-summary to review agent decisions.

Available Workflows

WorkflowPurpose
sanitize-corpusStrip instruction-like patterns from target content.
scan-untrusted-contentDetect prompt injection, hidden instructions, and exfiltration patterns.
classify-injection-riskScore content by injection danger level.
guardrail-checkValidate AI tool call against safety policy.

Evidence Required

  • Sanitized content comparison (before/after).
  • Injection pattern matches with context.
  • Tool call decision log with approval status.

References

  • OWASP LLM01:2025 (Prompt Injection)
  • OWASP LLM Prompt Injection Prevention Cheat Sheet
  • OWASP AI Agent Security Cheat Sheet
  • PortSwigger Web LLM Attacks Academy Topic

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.