agent-creator
Use when creating expert agents. Generates agent.md with frontmatter, hooks, required sections, and skill references.
Use when checking a specific dependency or package version for known CVEs and security advisories.
Claude Code
$ npx -y skills add fusengine/claude-plugins --skill cve-research --agent claude-code
Codex
$ npx skills add fusengine/claude-plugins --skill cve-research -a codex -y
Cursor
$ npx skills add fusengine/claude-plugins --skill cve-research -a cursor -y
Gemini CLI
$ npx skills add fusengine/claude-plugins --skill cve-research -a gemini-cli -y
GitHub Copilot
$ npx skills add fusengine/claude-plugins --skill cve-research -a github-copilot -y
OpenCode
$ npx skills add fusengine/claude-plugins --skill cve-research -a opencode -y
Windsurf
$ npx skills add fusengine/claude-plugins --skill cve-research -a windsurf -y
name: cve-research description: "Use when checking a specific dependency or package version for known CVEs and security advisories (OSV, GitHub Advisory, NVD), with CVSS, CWE and reachability verification before anything is reported as a finding." argument-hint: "<package-name> [version] [ecosystem]" user-invocable: true
<objective> This skill researches known vulnerabilities for one dependency at the version actually installed, across OSV.dev, the GitHub Advisory Database and NVD, then a web search for advisories not yet indexed. Each candidate is scored (CVSS vector + score + CVSS version + scoring source, CWE) and must pass two mandatory checks before it becomes a finding: the affected version is the one installed, and the vulnerable code is reachable from the project.
It reports; it never upgrades the dependency. Out of scope: a full project dependency sweep (use dependency-audit). </objective>
| Order | Source | Endpoint | Strength | |-------|--------|----------|----------| | 1 | OSV.dev | `api.osv.dev/v1` | Package + version matching, aliases (GHSA, CVE, RUSTSEC, GO, PYSEC) | | 2 | GitHub Advisory | `api.github.com/advisories` | Reviewed ranges, first patched version, CVSS v3/v4, CWE, EPSS, malware type | | 3 | NVD | `services.nvd.nist.gov/rest/json/cves/2.0` | NVD and CNA CVSS metrics, CWE, CISA KEV fields | | 4 | Web search | fuse-browser `browser_fetch` / Exa when available | Advisories not yet indexed, maintainer notes |
Endpoints, bodies and rate limits: [CVE APIs Reference](references/cve-apis.md). Ready-to-run requests: [Query Templates](references/templates/cve-query.md).
1. **Resolve the installed version** from the lockfile (not the manifest range). 2. **Query** OSV (`/v1/query` with ecosystem + name + version), collect IDs and aliases. 3. **Enrich** each ID: GitHub Advisory (CVSS, CWE, patched version), NVD (CVSS by source, KEV). 4. **Search** the web for the package name plus the current year, computed at run time (`date +%Y`), never a year written in this skill. 5. **Score**: CVSS vector + base score, the CVSS version, and who scored it; CWE id. 6. **Verify (mandatory)** — [Severity and Reachability](references/severity-and-reachability.md): version check, reachability check, refutation pass. Confidence 1-10; only ≥ 8 is a finding. 7. **Report** with fix version or workaround as an option for the owner.
The band is an input. The reported severity follows the plugin's single scale in `rules/security-rules.md`, decided after the reachability check.
| CVSS base score | Band | |-----------------|----------| | 9.0 - 10.0 | CRITICAL | | 7.0 - 8.9 | HIGH | | 4.0 - 6.9 | MEDIUM | | 0.1 - 3.9 | LOW |
When sources disagree (e.g. NVD scores v3.1 HIGH while the CNA scores v4.0 CRITICAL), report both with their source and use the highest for ordering. A CVE listed in CISA KEV (`cisaExploitAdd` in the NVD record) is noted as exploited in the wild.
id (GHSA/CVE + aliases), title, severity, confidence 1-10, CWE, standard reference (OWASP Top 10:2025 A03 Software Supply Chain Failures for a vulnerable dependency), package@version, evidence (advisory, affected range, installed version, reachability), exploit scenario, recommendation (fixed version / workaround).
A plugin ecosystem that turns Claude Code into a supervised, multi-agent development environment.
Use when creating expert agents. Generates agent.md with frontmatter, hooks, required sections, and skill references.
Use when starting ANY development task -- feature, bug fix, refactor, hotfix (triggers: implement, create, build, fix, add…
Use when creating a feature/component or adding functionality. Fires BEFORE APEX Analyze to refine requirements via structured…
Use before a root-cause, done/verified claim, irreversible action, or 2nd-time fix reaches the owner (APEX or plain…
Use when validating code quality after modifications -- SOLID compliance, DRY duplication, linter errors, architecture…
Use when an expert agent self-reviews and self-corrects code after the Execute phase, before sniper validation (BMAD-METHOD…
You MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or…
Ask which skill or flow fits your situation. A router over the skills in this repo.