Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/nathan-oakes-build
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: testing, devops
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions: not checked yet.
What's inside
- Skill name
build- Description
- AIDLC Build + Test orchestrator (TDD). Delivers an open PR with green CI per Tech Spec; re-enters to triage /review PR comments — fix or reply+resolve. Not for spec-only work.
- License
- MIT
- Instructions
- 50 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# /build — Build + Test (phase orchestrator) You are the **phase orchestrator** for AIDLC **Build** and **Test** as **one practice**: tests are written **with** the code (TDD), not in a separate follow-up stage. Canonical definitions: - **AIDLC:** `docs/AIDLC.md` at the repository root — Build phase, Test phase, V-model. **Library skills:** [docs/SKILLS.md](../../docs/SKILLS.md). Resolve bundles from your install or `.claude/skills/<bundle>/` in the workspace. ## Inputs - Approved `feature/<slug>/tech-spec.md` - **If re-entering after `/review`:** open **PR** with **AIDLC Review — …** comments (see `/review` orchestrator). ## Orchestration — initial implementation 1. **Branch:** use a descriptive branch (e.g. `feature/<slug>-short-name`). Apply **`git-workflow`** ([skills/git-workflow/SKILL.md](../git-workflow/SKILL.md)). 2. **Implement by Tech Spec section:** in PR/commits, reference which section you are implementing (AIDLC Build guidance). 3. **TDD:** for each unit of work, prefer **test first or alongside** — frontend (`npm test` / vitest as applicable), backend (`dotnet test`, etc.). Load **`testing`** ([skills/testing/SKILL.md](../testing/SKILL.md)); use **`frontend-web`** for UI, **`backend-saas`** for API layers. 4. **Do not** “finish code” and add tests only at the end unless the Tech Spec explicitly sequenced an exception. 5. **Open a PR** targeting the repo’s integration branch (e.g. `main` / `develop` per **`git-workflow`**). **Build is not complete** until the PR is **open** and **CI is green** (required checks passing on the latest commit). Iterate until green — same bar as AIDLC Phase 3 in `docs/AIDLC.md`. 6. **Local checks** first, then push; treat remote CI as authoritative for handoff to Test/Review. ## Review feedback loop (after `/review` has posted on the PR) When **`/review`** has run, each **dimension** (Tech Spec, Testing, DevOps, Frontend/UX, Security) should have left **GitHub PR comments** (preferred). The **build** orchestrator **owns the response**: 1. **Read** all open **review threads** on the PR — especially comments titled `AIDLC Review — …`. 2. **For each finding** (or each thread), decide: - **Valid:** implement the fix (code/tests/config/docs as appropriate), push commits, and **reply** on the same thread briefly stating what changed **or** mark the conversation **resolved** once the fix is on the branch (per team habit). - **Invalid / won’t fix (with cause):** **reply** on the **same GitHub comment thread** with a **clear rationale** (cite Tech Spec section, intentional scope, or false positive). Then **resolve the conversation** so reviewers see closure. 3. **Do not** silently ignore review feedback — every thread gets either a **code change** or a **documented reply**. 4. Re-run **local build/tests**; ensure **CI** is green. 5. If changes were substantive, run **`/review`** again for a **follow-up pass**; otherwise proceed toward merge per team rules. **Tools:** use **GitHub MCP**, **`gh api` / `gh pr comment`**, or web UI instructions for the human if the agent cannot post — but **prefer** direct PR replies. ## Nested library skills (typical) | When | Skill | |------|--------| | Implementation patterns | `frontend-web`, `backend-saas`, `architecture` | | Tests | `testing` | | Commits / PR | `git-workflow` | ## Outputs - **Initial Build pass:** **Open PR** + **green CI** + code/tests implementing the Tech Spec; traceability to Tech Spec sections in commits/PR body. - **After `/review`:** same PR with **addressed or replied-to** review threads, CI still green.
From nathan-oakes/AI-DLC, MIT.
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Apr 13, 2026 (6 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/nathan-oakes/AI-DLC
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
Same skill elsewhere
Copies of this SKILL.md published by others; search shows only one of them.
Cite this page
build skill. Agent Mart. https://agentmart.dev/skills/nathan-oakes-build (last updated 2026-10-24)