Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add tranfu-labs/tranfu-skills --skill prd-to-parallel-loopgit clone --depth 1 https://github.com/tranfu-labs/tranfu-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/tranfu-labs/tranfu-skills/prd-to-parallel-loop)<a href="https://agentmods.dev/skills/tranfu-labs/tranfu-skills/prd-to-parallel-loop"><img src="https://agentmods.dev/badge/skills/tranfu-labs/tranfu-skills/prd-to-parallel-loop/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/tranfu-labs/tranfu-skills/prd-to-parallel-loop"><img src="https://agentmods.dev/badge/skills/tranfu-labs/tranfu-skills/prd-to-parallel-loop.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00451 | $0.08607 |
| Opus 5 | $0.00226 | $0.04304 |
| Sonnet 5 | $0.00090 | $0.01721 |
| Haiku 4.5 | $0.00045 | $0.00861 |
Grade A, and why
prd-to-parallel-loop scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 420 lines — stays where its author put it; the contents beside it link to each section on GitHub.
prd-to-parallel-loop
这个 skill 做什么
把一批要并行开发的变更搭成可跑的 loop 脚手架。核心对应关系:一个 openspec change = 一个并行任务 = 一个 git worktree = 一个 codex 会话。
PRD / 产品文档 或 里程碑文档 (docs/roadmap/milestone-N-xxx.md)
↓ 切分与依赖协商(里程碑文档里已经写好就沿用,不重写)
openspec/changes/<change-id>/ × N ← 变更单元 = 并行单元
↓ 一个 change 一个 worktree
~/dev-loops/<project>/<slug>/runtime.json ← 依赖 DAG 与调度状态的事实源
↓ /loop 心跳
主 loop 分三段驱动每个任务:实施 → 反思 → 归档 → 自己 squash merge 落地
跑完 skill,用户在新对话粘贴 loop-prompt.md 里那段 + /loop 就能开跑。
为什么是 change 粒度,不是 change 内的 task 粒度
change 是 openspec 里"一个连贯的变更提案",也是天然的可归档单元。把一个 change 切成 N 份丢给 N 个互相看不见的 codex,等于把提案的内部一致性交给并发去赌,还会让 N 个执行者共写同一份 tasks.md 与 spec-delta。按 change 并行则每个 worktree 自成闭环:codex 在里面走完 openspec-driven-development 的步骤 5→7,产出的 spec-delta 走标准路径,归档动作也在它自己手里。
任务的三段式生命周期
主 loop 每次心跳只推进一段,段与段之间是调度器的检查关:
| 段 | 谁做 | 内容 |
|---|---|---|
implement |
codex(spawn) | 按 change 定义写码,勾自己的 tasks.md,写 spec-delta |
reflect |
codex(resume) | 对照 proposal/design/tasks 核有无遗漏与有无多做,有则改 |
archive |
codex(resume,已 rebase 到最新 main 之后) | spec-delta 合进 specs/、change 移进 archive/ |
| 落地 | 主 loop(不占独立心跳) | squash merge → 核对归档结果 → 在合并后的工作区跑验证 → 勾里程碑那一行 + 写证据锚 → commit |
rebase 卡在 archive 之前,是因为归档要写 openspec/specs/<domain>/spec.md——多 worktree 并行时唯一的共享写点。让它在最新 main 上做,冲突提前摆到 codex 面前解掉。
落盘范围(ownership: edit-file)
本 skill 会直接写入用户文件系统。写盘范围仅限:
{{PROJECT_REPO}}/openspec/changes/<change-id>/× N(从项目openspec/changes/_template/拷贝再填)~/dev-loops/bin/下的 3 个.mjs(已存在则跳过,永不覆盖)~/dev-loops/<project-name>/<slug>/下的 runtime.json / loop-instructions.md / loop-prompt.md /(必要时)acceptance-map.md
绝不写用户的里程碑文档或 PRD。里程碑文档是人类文档,skill 只读它;唯一的写入发生在 loop 跑起来之后、由主 loop 在落地时勾对应那一行 checkbox 并追加证据锚——那是运行时行为,不是本 skill 的落盘。
也绝不修改 openspec/changes/_template/、git 历史、产品代码,或已在跑的 runtime.json。
这个 skill 不做什么
- 不代替产品文档写作。PRD / 里程碑文档怎么写、验收怎么定,由用户与 credibility-review / project-init-docs 负责。尤其不给里程碑文档规定格式——skill 尽力解析,解析不到就问
- 不代替 codex 里的实现判断。怎么写代码、用什么工具验证,是 codex 的事;skill 只保证它拿到清晰的 change 定义、明确的验证命令、清楚的分段规范
- 不建 openspec 骨架。项目缺 openspec/ 就打回,让用户先走 openspec-driven-development
- 不接单个 change。拆出来只有一个 change 就打回:不值得起 loop,直接走 openspec-driven-development
- 不推送、不合并到远端。主 loop 只本地 squash merge 到 main;push 是用户人工决定
- 不做 update 模式。runtime.json 已在跑就拒绝重建;强行覆盖会丢失 phase / attempts / merged_sha / codex_session_id 等运行时数据
What ships with it
11 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- agents/openai.yaml 80 B
- assets/bin/probe-codex.mjs 4.2 KB runs code
- assets/bin/resume-codex.mjs 4.3 KB runs code
- assets/bin/spawn-codex.mjs 4.7 KB runs code
- assets/icon.png 691 B
- assets/icon.svg 470 B
- assets/templates/dev-loops/loop-instructions.md.tmpl 24 KB
- assets/templates/dev-loops/loop-prompt.md.tmpl 2.1 KB
- assets/templates/dev-loops/runtime.json.tmpl 1.3 KB
- README.md 4.9 KB
- README.zh.md 4.2 KB
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 420 lines · 451 tokens per session scan A d385403986b9
prd-to-parallel-loop is a skill published in the GitHub repository tranfu-labs/tranfu-skills (2 stars, last pushed 2d ago), licensed MIT. It adds 451 tokens to every session and 8,607 once invoked, about $0.0023 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
prowler-commit
Creates professional git commits following conventional-commits format. Trigger: When creating commits, after completing code changes, when user asks to commit.
gh-auth-isolation
Safely manage multiple GitHub identities (EMU + personal) in agent workflows.
comet-github
A routing guide for Comet-related GitHub work. It directs requests about pull requests, issues, CI failures, ideas, and fixes to the appropriate review or implementation process.
github-skill
Work with GitHub via the gh CLI — clone repositories, create/list/merge pull requests, create/list issues, and run any other gh command (API calls, workflow runs, releases, repo administration). List operations return parsed JSON.
changelog-composer
Generates structured changelogs and release notes from git history and PRs, classifying breaking changes, features, fixes, performance, docs. Triggers on: "generate changelog", "write release notes", "what changed since", "prepare release", "release notes for", "diff since tag".
re0-merge
Review and land an external contribution the way this suite does: gate it against the thesis, land it with the author's credit intact, complete a new skill rather than merging it raw, then approve, credit, and explain before closing. Use when reviewing a pull request, as any collaborator or maintainer, not only the…