agent-team-generator

A Claude Code skill that scaffolds a complete, disciplined AI agent team for your project:

• an interview that extracts your stack, risk surfaces, and business rules — including which harness (Claude Code, OpenCode 2, both kept in line, or another), whether you want separate backend + UI/UX developers (parallel dispatch) or a single full-stack developer (simpler roster), and how the main agent should report to you (technical / direct / plain English — always outcome-first and structured); a fast path accepts the recommended defaults in one question
• AGENTS.md + CLAUDE.md foundation files
• specialist subagents for both Claude Code (.claude/agents/) and OpenCode 2 (.opencode/agents/, provider/model#variant pins, permissions as a rule list in opencode.json), each with a stated lane and per-project non-negotiables
• a Fable-level process protocol (.agents/rules/claude-agent-protocol.md) so mid-level models follow frontier-model steps — the quality bar lives in the process files, not the model — plus a language-agnostic engineering standard with per-stack hygiene rules (TypeScript, Python, Go, Rust)
• the senior ladder, in every agent that writes code: does it need to exist → is it already in the codebase → does an installed dependency do it → can it be one line → only then the minimum that works — with a floor that never drops error handling, graceful user-facing errors, or user experience
• a Design bar in every agent that builds UI — a design direction before any pixels, tokens, designed states, whole pages, and a list of generated-looking patterns to redo — plus recommended design skills (frontend-design, Impeccable, UI/UX Pro Max)
• an independent code-reviewer you run on request: it reviews the diff and the code around it, grades findings 🔴 Blocker / 🟠 Should fix / 🟡 Nit / 🔵 FYI / 🟣 Minor, proves the tests can fail by changing a value, logs minor issues to documentation/known-issues.md, and hands back a ready-to-run fix prompt
• open permissions, user-gated actions: by default agents may edit any file and run any command, so an instruction is carried through without prompts — and they commit freely but push, open PRs, deploy, or destroy anything only when you say so. Stricter tiers (Open, Guarded, Standard, Strict) add a mechanical net, written once to opencode.json and .claude/settings.json
• a mechanical verifier (.agents/verify-team.js) left in your repo: placeholders, roster vs files, OpenCode 2 shape, permission parity across harnesses, gate scripts
• a plain-English documentation/ system maintained as part of Definition of Done
• optional third-party skill add-ons (Animation Principles, Anthropic's web-app testing, Superpowers) — offered during the interview, installed at project or user level, your choice — plus a security-review command for OpenCode parity with Claude Code's built-in /security-review

There is no project-manager agent. The main agent — the session you are talking to — leads: it dispatches the specialists (product-specialist, architect, developers, qa-tester) and, for small quick changes, does the work itself under that specialist's rules.

Install

npx agent-team-generator

That copies the skill to ~/.claude/skills/agent-team-generator, making it available in every project. To install for a single repo instead:

npx agent-team-generator --project

Other flags: --force (overwrite an existing install), --uninstall, --help.

To upgrade to the latest version:

npx agent-team-generator@latest --force

Or install as a Claude Code plugin (no Node required) — inside Claude Code run:

/plugin marketplace add sams-git-195/agent-team-generator
/plugin install agent-team-generator@agent-team-generator

Use

Open Claude Code in the project you want an agent team for and run:

/agent-team-generator

…or just say "set up my agent team". The skill will survey the repo, interview you phase by phase, propose a roster for approval, and only then generate files. Unknowns become explicit ⚠️ undecided markers — never guesses.

Repo layout

| Path | What | |---|---| | skills/agent-team-generator/SKILL.md | The skill entrypoint (workflow, steps, verification) | | skills/agent-team-generator/references/ | Interview bank, protocol + AGENTS.md templates, role library, Fable playbook, engineering + design standards, permission policy | | skills/agent-team-generator/references/templates/ | Pre-filled agent files for the core roles (incl. code-reviewer) + fullstack-developer | | skills/agent-team-generator/scripts/verify-team.js | Mechanical post-generation checks; copied into the target's .agents/ | | skills/agent-team-generator/fixtures/ | Canned interview for testing the skill itself | | bin/cli.js | The npx installer | | .claude-plugin/ | Plugin + marketplace manifests for Claude Code's /plugin install path | | scripts/test.js | Smoke test (npm test): installer round-trip + template integrity | | .github/workflows/ | ci.yml (tests on PR + merge) and release.yml (publish on tag) |

Releasing

Publishing runs from CI via npm trusted publishing — no npm token is stored in the repo or in GitHub secrets. To cut a release, bump the version in both package.json and .claude-plugin/plugin.json (the smoke test fails if they disagree), then:

npm version patch && git push --follow-tags

The tag triggers release.yml, which checks the tag matches package.json, runs the tests, and publishes with provenance.

One-time setup on npmjs.com (after the first manual publish): package → Settings → Trusted Publisher → GitHub Actions, with organization/user sams-git-195, repository agent-team-generator, workflow filename release.yml, no environment. Fields are case-sensitive.

License

MIT