Borrowing it
Nothing to install: this file belongs to yusifeng/formax. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/yusifeng/formax/main/.codex/skills/formax-dev-loop-workflow/SKILL.mdgit clone --depth 1 https://github.com/yusifeng/formaxWrote 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/yusifeng/formax/formax-dev-loop-workflow)<a href="https://agentmods.dev/skills/yusifeng/formax/formax-dev-loop-workflow"><img src="https://agentmods.dev/badge/skills/yusifeng/formax/formax-dev-loop-workflow/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/yusifeng/formax/formax-dev-loop-workflow"><img src="https://agentmods.dev/badge/skills/yusifeng/formax/formax-dev-loop-workflow.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00068 | $0.00628 |
| Opus 5.5 | $0.00027 | $0.00251 |
| Sonnet 5.5 | $0.00014 | $0.00126 |
| Haiku 4.5 | $0.00007 | $0.00063 |
Grade A, and why
formax-dev-loop-workflow 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 2mo 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 — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Formax Dev Loop (Mainline Discipline)
Default loop (repeat per TODO item)
-
Pick one mainline item and finish it end-to-end before starting another.
- If a new idea appears mid-flight, write it down in a backlog note and continue the mainline.
-
Write/adjust tests first to lock behavior (when feasible).
-
Implement the smallest change that satisfies the item.
-
Incremental optimization check (required)
- After implementation, do a quick pass:
- Is there newly-dead/unused logic introduced by this increment?
- Is there a low-risk simplification that reduces branching/duplication?
- If yes, include a small optimization in the same item (no scope drift, no behavior change).
- If no, explicitly proceed without optimization.
- After implementation, do a quick pass:
-
Run only targeted tests (never
bun run test:coverageunless explicitly asked).- Preferred:
bun run test -- <changed-test-files...> - Helper (repo):
bun run test:changed- Use default (staged only) for the commit you are about to make.
- Use
bun run test:changed -- --allonly when you intentionally want staged + unstaged + untracked.
- Preferred:
-
Pre-commit hygiene
- Avoid partial staging (“MM” state). If needed, check with:
bun run check:partial-stage
- Run review before every commit using
AGENTS.md->Review Profile (Single Source of Truth). - If review returns findings: fix -> re-run targeted tests -> re-run review.
- Avoid partial staging (“MM” state). If needed, check with:
-
Commit
- Keep it small (2–4 files ideally, unless refactor forces more).
- Prefer one concern per commit (tests + implementation together for that concern).
Guardrails (Formax-specific)
- Do not fix unrelated failures mid-loop unless they block the mainline.
- If a command would usually be run with pipes/redirections for convenience, prefer running it plain and rely on Formax’s own output truncation/Expand UI.
- Don’t “clean up” formatting/copy/colors/spacing unless explicitly requested or required for parity.
Quick commands
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.
- 2mo ago First seen · 62 lines · 68 tokens per session scan A 0b7f03f259df
formax-dev-loop-workflow is a skill published in the GitHub repository yusifeng/formax (184 stars, last pushed 3mo ago), licensed MIT. It adds 68 tokens to every session and 628 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-08-30.
formax-dev-loop-workflow compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| formax-dev-loop-workflowyusifeng/formax | Skill | 184 | 3mo ago | MIT | A | 68 | Codex |
| autofix-prQwenLM/qwen-code | Skill | 29k | yesterday | Apache-2.0 | A | 141 | Claude Code |
| githubopenclaw/openclaw | Skill | 394k | yesterday | MIT | A | 28 | Claude Code, Codex |
| cw-landHmbown/Codewhale | Skill | 41k | 1mo ago | MIT | A | 44 | Claude Code, Codex |
| git-commitUfoMiao/zcf | Skill | 6.1k | 8d ago | MIT | A | 41 | Claude Code |
| git-mastercode-yeongyu/lazycodex | Skill | 3.9k | yesterday | MIT | A | 93 | Codex |
| atmos-gitcloudposse/atmos | Skill | 1.4k | yesterday | Apache-2.0 | A | 82 | Claude Code |