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 skills/markd1zzz/workflow-tracker/workflow-trackernpx skills add MarkD1Zzz/workflow-tracker --skill workflow-trackergit clone --depth 1 https://github.com/MarkD1Zzz/workflow-trackerWhat 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 | $0.00100 | $0.01793 |
| Opus 5 | $0.00050 | $0.00897 |
| Sonnet 5 | $0.00020 | $0.00359 |
| Haiku 4.5 | $0.00010 | $0.00179 |
Grade A, and why
workflow-tracker 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 2d 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 — 178 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Workflow & Experiment Tracker
自动拆解为 Phase → Task → Experiment 三级结构。识别项目类型后输出对应格式。
触发
检测到以下任一信号即静默记录，不打断用户：
- 看到数字变化（"精度从X变到Y"、"loss降到Z"、"F1从A变到B"）
- 用户口述实验结论（"试了X"/"不行"/"有效"/"放弃"）
- 运行脚本后看到结果输出
- 参数变更（"把A改成B"、"新增了C"）
- 用户说"记录"/"保存"/"更新日志"
不必等到用户说"记录"才行动。在回复中自然地追加一条记录即可。
项目类型识别
首次进入项目时，扫描目录+用户对话，判定类型：
| 信号 | 类型 | 存储 |
|---|---|---|
含 tex/、manuscript、figures/、投稿、论文、ablation |
论文 | CHANGELOG.md + experiment_log.md |
含 data/train/、models/、train.py、pipeline、main.py |
工程 | workflow.json + workflow.md |
| 用户说"部署"/"产品"/"上线"/"API" | 工程 | workflow.json + workflow.md |
| 无法判断 | 默认工程 | workflow.json + workflow.md |
论文模式
直接追加两个 Markdown 文件，不建 JSON：
CHANGELOG.md
按时间线累积。每条格式：
## 2026-06-14 — 简短标题
### 背景
（1-2句说明为什么做这个改动）
### 改动
| 参数 | 旧值 | 新值 | 原因 |
|------|------|------|------|
### 效果
（关键数字变化）
experiment_log.md
每次实验一条，可独立阅读：
## 2026-06-14 — 实验标题
- **假设**: ...
- **方法**: ...
- **参数**: ...
- **结果**: 基线X → 新Y (delta)
- **结论**: SUCCESS / FAILED / PARTIAL
- **备注**: ...
工程模式
维护 .claude/workflow/workflow.json + 自动导出 .claude/workflow/workflow.md。
JSON 最小结构
{
"project": "项目名",
"updated": "2026-06-14T14:30",
"phases": [{
"name": "Phase 1: XXX",
"status": "in_progress",
"tasks": [{
"name": "Task 1.1: XXX",
"status": "completed",
"experiments": [{
"date": "2026-06-14",
"title": "实验标题",
"method": "一句话方法",
"params": {"old_lr": 0.001, "new_lr": 0.0002},
"results": {"baseline": 93.75, "new": 94.79, "delta": 1.04},
"conclusion": "SUCCESS",
"tags": ["classifier", "svm"]
}]
}]
}]
}
字段说明：
title/method/results/conclusion必填，其他可选params用 old_xxx / new_xxx 格式记录变更results.delta统一用 pp（百分点）或绝对值tags至少一个分类标签conclusion取值:SUCCESS | FAILED | PARTIAL | INCONCLUSIVE
workflow.md 自动导出
每次更新 JSON 后同步重写。格式：
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.
- 2d ago First seen · 178 lines · 100 tokens per session scan A 18753d7b7773
workflow-tracker is a skill published in the GitHub repository MarkD1Zzz/workflow-tracker (10 stars, last pushed 2mo ago), licensed MIT. It adds 100 tokens to every session and 1,793 once invoked, about $0.0005 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
superpowers-graph
Runs the Superpowers software-development methodology as a guarded state machine — brainstorm, spec, plan, isolated workspace, the per-task implement/review/fix loop, the whole-branch review, and the integration menu — with typed node contracts, total exit guards, bounded retry loops, and a ledger recording every step…
plan-build-review
Rigorous think→plan→do→review workflow. Use when the user invokes /plan-build-review, /plan-build-review:plan-build-review, or asks for a planned and reviewed implementation with optional git worktree, commit, push, or deploy finishing.
html-ppt-hermes-cyber-terminal
OpenDesign + BYOK: choosing and wiring your own model, hands-on — cost, quality, and the routing decision. Built as a decision-grade AI literacy deck for engineers, IT, applied-AI teams.
development
开发语言能力索引。Python、Go、Rust、TypeScript、Java、C++、Shell。当用户提到编程、开发、代码、语言时路由到此。.
post-build-flow
Handles workflow verification and setup after build-workflow succeeds, or when the message contains workflow-verification-follow-up or workflow-setup-required. Load after direct builds, when verificationReadiness requires action, or on orchestrator verify/setup follow-up turns.
n8n:create-pr
Creates GitHub pull requests with properly formatted titles that pass the check-pr-title CI validation. Use when creating PRs, submitting changes for review, or when the user says /pr or asks to create a pull request.