Audit or rewrite AGENTS.md so it holds only lasting principles. Run it only when the user asks for it by name; never invoke it on your own.
中文简介正在准备中。当前显示原文说明。
dzhng/skills13742026年10月19日 更新
免费在 GitHub 公开
implement-spec-with-codex
Run [implement-spec](../implement-spec/SKILL.md) with Codex doing the implementation passes while you orchestrate, integrate, and review. Use when the user asks to implement a spec with Codex, or to run implement-spec with Codex as the implementer.
中文简介正在准备中。当前显示原文说明。
查看安装方法包含的文件（1）
- SKILL.md2.1 KB
SKILL.md（原文）
安装前可以查看交给 AI 代理的指令内容。
Implement Spec with Codex
Run implement-spec exactly — same loop, same
gates, same maintenance checkpoints — but Codex writes the code. Invoking this
skill IS the explicit ask that authorizes codex exec delegation under the
codex skill.
Role split
- Codex implements — whenever possible. Default every implementation pass
to a
codex exec, sliced sharp and prompted per the codex skill (prompt blocks, sandbox choice, exec liveness). Run independent slices as concurrent execs in separate worktrees, exactly where implement-spec would fan out subagents. - You orchestrate and review. The spec loop stays yours: pick the next slice, reconcile plan with code, author each prompt end-to-end, integrate diffs, run the gates, review the merged tree, commit, update the handoff, run maintenance checkpoints. None of that delegates.
Keep a pass yourself only when slicing it sharply would cost more than doing it — tiny fixups, plan reconciliation, integration conflicts.
Rules
- You own every result: read the full diff and run the verification yourself. "Codex says it's done" is never done.
- Visual results carry implement-spec's two extra proofs (production-route pixel diff + unprimed critique) with double force here: a sandboxed Codex never saw its change render, so a pass that misses the production render path entirely arrives looking fully "verified".
- A sandboxed Codex ships code it never saw run — browser-verify every visual slice yourself and budget fix rounds. Send a red gate back as a resumed follow-up with the failing evidence; fix it yourself only when that's smaller than re-prompting.
- Commit hygiene stays yours: review the change list of the merged tree before every commit, exactly as implement-spec demands of the integrating reviewer.
评价
还没有评价。欢迎分享你的使用感受。
评论
无需登录即可发表。欢迎分享使用感受、提问，或说明在哪些环境下无法运行。
还没有评论。来写第一条吧。
同一仓库的技能
Audit the choices an implementing agent made, not its diff — a pure decision audit that traces the session's history into a choices ledger, changes no code, and never blocks an unsupervised run. Working code still embeds architecture the user never chose; surface it because future work inherits it. Use when the user wants to review the decisions the AI made on their behalf, before merging or committing AI-implemented work, when integrating a delegated subagent's pass, or when a fix "works" but might be a point fix.
中文简介正在准备中。当前显示原文说明。
dzhng/skills13742026年10月19日 更新
Audit and prioritize performance work through bounded-work and forward-progress checks. Use when asked to find performance issues, investigate freezes/500s/OOMs, review retries or polling for no-progress loops, rank a performance backlog, or implement the next simple performance fixes.
中文简介正在准备中。当前显示原文说明。
dzhng/skills13742026年10月19日 更新
Audit whether tests earn their maintenance cost and which suite owns each contract. Use when pruning redundant tests, investigating implementation coupling or test-only production hooks, reviewing the value of proposed coverage, or auditing an entire subsystem. Use write-tests to implement the resulting test changes.
中文简介正在准备中。当前显示原文说明。
dzhng/skills13742026年10月19日 更新
Run fast, progressive experiments to map tunable parameters and their effects. Use when optimizing code, prompts, configurations, or other artifacts against a goal or benchmark, or when a long research run needs a planned hypothesis queue, focused trials, and durable findings.
中文简介正在准备中。当前显示原文说明。
dzhng/skills13742026年10月19日 更新
Shoot software UI like a film. A sequence of hero shots, each a steep, oblique, extreme close-up of the interface floating in black, with a razor-thin focal plane, a slow motivated camera drift, and UI that acts on its own, hard-cut on the action, ending on one line of copy and a still logo. Use when asked for a launch video, product film, feature trailer, teaser, promo clip, hero shot, or any "cinematic" footage of an app or UI. This skill covers camera, framing, light, and edit. It is stack-agnostic.
中文简介正在准备中。当前显示原文说明。
dzhng/skills13742026年10月19日 更新