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 agentmods add commands/touheedcode/claude-dev-workflow/retrogit clone --depth 1 https://github.com/TouheedCode/claude-dev-workflowWhat 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 | $0.00000 | $0.00992 |
| Opus 5 | $0.00000 | $0.00496 |
| Sonnet 5 | $0.00000 | $0.00198 |
| Haiku 4.5 | $0.00000 | $0.00099 |
Grade A, and why
retro 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 yesterday.
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 — 123 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/retro — Feature Retrospective
Arguments
$ARGUMENTS— Optional: story ID (e.g., "US-001") or branch name. If omitted, analyzes the current branch.
Instructions
You are generating a retrospective report for a completed feature. Your job is to analyze git history and project artifacts to surface patterns, measure delivery quality, and suggest workflow improvements.
Input: $ARGUMENTS
Step 1: Gather Data
Git History
# All commits on the feature branch
git log main..HEAD --oneline --format="%h %s"
# Commits per phase (by scope in commit message)
git log main..HEAD --oneline --grep="(ui):" --grep="(api-client):" --grep="(smart-components):" --grep="(repository):" --grep="(service):" --grep="(controller):" --grep="(integration):"
# Files changed per commit
git log main..HEAD --stat --format="%h %s"
# Total diff stats
git diff main..HEAD --stat
Project Artifacts
- Read
.phase-context.jsonfor the story ID and completed phases - Read the plan at
docs/plans/PLAN-{story-ID}.mdfor planned vs. actual deliverables - Read the user story at
docs/stories/{story-ID}.mdfor BDD scenarios
Step 2: Analyze Phase Delivery
For each phase (1–7), report:
| Metric | How to Measure |
|---|---|
| Commits | Count commits matching the phase scope |
| Files created | Count new files in that phase's commit(s) |
| Files modified | Count files changed after initial commit (rework) |
| Test count | Count it() or test() calls in test files |
Flag phases with high rework (multiple commits modifying the same files), which indicates the phase wasn't completed cleanly on the first pass.
Step 3: Identify Patterns
Analyze the git history for these patterns:
Iteration Patterns
- Single-pass phases — completed in one commit (ideal)
- Multi-commit phases — required fixes after initial implementation
- Fix-after-validate — commits that follow a validation failure
Common Issues
Look for these in commit messages and diffs:
- Type errors fixed after implementation
- Test failures fixed after writing
- Lint fixes as separate commits
- Import/export issues between shared types and consumers
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.
- yesterday First seen · 123 lines · 0 tokens per session scan A f264b6a6927c
retro is a command published in the GitHub repository TouheedCode/claude-dev-workflow (2 stars, last pushed 2d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 992 tokens. 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 commands, from other repositories
auto-work
给一个需求,AI自动完成调研→方案→方案Review→开发→开发Review全流程.
develop-review
Review feature-developing 生成的代码,检查遗漏和宪法违规.
developing
Command "developing" from chaohong-ai/ai-auto-work, covering 参数解析, 你的角色, 工作流程, 第一步:建立完整上下文 and 第二步:确认实现范围.
manual-work
带人工检查点的开发流程:前期需求/方案重点把关,后期自主执行.
plan-review
Review生成的plan,检查边界情况、场景覆盖、架构合理性.
analyze
分析 auto-work 产出中遗漏/缺陷的根因,定位流程断点并输出修复方案+工作流优化建议.