superagent-aiOfficial·13 Agent Skills Included

skills

Offline security audits for code, dependencies, CI/CD, and infrastructure

Audits application code, dependencies, GitHub Actions workflows, and infrastructure files for real security risks like broken authorization, leaked secrets, supply-chain attacks, and cloud misconfigurations. Replaces slow manual reviews and external scanners with offline checks that report each issue at file:line with a concrete fix. Covers pentest planning, vulnerability triage, and skill safety vetting so teams ship hardened software faster.

76|11

↗

npx skills add superagent-ai/skills --all -g -y

All Skills in This Repository (13)

Pure Emerald Level Indicators

📦 In Repo

superagent-ai

skill-security

Audit AI skills for safety using deterministic scanning and contract checks.

authz-security

Identify missing authorization checks in routes, controllers, resolvers, and data models.

repo-security-posture

Inventory GitHub repository security posture across branch protection, actions, and access controls.

security-disclosure-triage

Evaluate security advisories against repository checkouts and assign severity.

supply-chain-security

Analyze manifests, lockfiles, and install scripts to flag supply-chain risks.

crypto-secrets

Scan application code and configuration for cryptography and secrets hygiene issues.

pr-github-ops

Post Superagent PR security scan findings as inline GitHub review comments via gh CLI.

redteam-autoresearch

Generate labeled adversarial prompts and responses for guardrail training data.

ci-cd-security

Identify security weaknesses in GitHub Actions workflow YAML files.

hacker

Orchestrate authorized offensive security engagements with phase gates and role subagents.

vulnerability-triage

Analyze vulnerability reports against project intent and produce a structured triage report.

recon-security

Coordinate and execute authorized external penetration tests from reconnaissance to reporting using free, open-source tools and structured evidence management.

Official

Advanced

Frequently Asked Questions

FAQPage Schema

How to install superagent-ai/skills?▼

Run `npx skills add superagent-ai/skills --all -g -y` in your terminal to install all security skills globally.

How to check if an AI skill is safe to install?▼

Use the skill-security skill, which runs an offline scanner plus a semantic review to catch prompt injection, credential theft, and hidden malicious code before you install anything.

Can it find broken access control in my code?▼

Yes. The authz-security skill reads your routes and controllers offline and flags IDOR, BOLA, and mass-assignment bugs with framework-correct fixes.

Does it work without cloud credentials or running apps?▼

Yes. Every audit is read-only against files on disk, so no credentials, deployments, or external services are required.

Can I use these skills without security expertise?▼

Yes. Each finding comes with a plain-language severity, an explanation of the risk, and a concrete corrected code snippet you can apply directly.