agentleFS
wedabrowedabro/bro-skills/.agents/skills/speckit.security/SKILL.md

speckit.security

Skill3 starsChanged 45 days agoMITView on GitHub

Security Auditor - Perform evidence-backed application, dependency, secret, and threat-model audits without silently modifying code.

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 3 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Read source, history scope approved for review, manifests/lockfiles, Docker/config, feature spec, constitution, and existing security …

    line 10
  2. Distinctive:

    Write a redacted security report with scope, commands/tools, findings, evidence locations, severity, confidence, remediation, owner, and …

    line 11
  3. Very specific:

    Run applicable checks: OWASP injection/XSS/SSRF/IDOR/authz, secret/config exposure, dependency/supply-chain, container/runtime, and …

    line 15
  4. Distinctive:

    Every applicable check has a result or explicit N/A reason, each finding has file/line or configuration evidence and confidence, and the …

    line 22
  5. Very specific:

    Read-only. Do not silently patch, delete data, invent CVEs, expose secret values, or provide harmful exploit code.

    line 25

≈ “Common” is an estimate: at least 1 in 5 of the 308,232 skill files we read have a rule of the same kind, not the same words. “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

Mission

Identify and explain security risk across code, configuration, dependencies, identity, data, and runtime boundaries, then produce actionable remediation and residual risk.

Inputs and Outputs

  • Read source, history scope approved for review, manifests/lockfiles, Docker/config, feature spec, constitution, and existing security reports.
  • Write a redacted security report with scope, commands/tools, findings, evidence locations, severity, confidence, remediation, owner, and residual risk.

Protocol

  1. Establish scope, revision, threat actors, assets, trust boundaries, data classes, and exclusions. Never send private code or secrets to an external service.
  2. Run applicable checks: OWASP injection/XSS/SSRF/IDOR/authz, secret/config exposure, dependency/supply-chain, container/runtime, and sensitive logging. Record tool versions and exit codes.
  3. Trace sensitive flows through authentication, authorization, tenant/ownership checks, validation, persistence, queues, webhooks, and external calls. Test denial paths, replay, rate limits, and failure handling where evidence is available.
  4. Validate findings: reproduce safely or cite a precise static path; distinguish confirmed vulnerability, likely risk, and missing evidence. Do not publish exploit payloads or echo secret values.
  5. Prioritize by impact × likelihood × confidence; include affected asset, trigger, blast radius, minimal fix direction, regression check, and remediation deadline only when supplied.
  6. Re-scan after an authorized fix and compare findings by stable ID. A closed finding needs evidence that the original trigger no longer succeeds; unresolved risk remains open.

Completion Gate

Every applicable check has a result or explicit N/A reason, each finding has file/line or configuration evidence and confidence, and the conclusion states limitations. Any Critical confirmed finding yields FAIL.

Guard Rails

Read-only. Do not silently patch, delete data, invent CVEs, expose secret values, or provide harmful exploit code. Defer remediation to fix-security-vulnerabilities-with-strix when that workflow is explicitly selected.

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.