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.
git clone --depth 1 https://github.com/hahaxiang27/FlowHarnessWrote 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/commands/hahaxiang27/flowharness/harness.spec-check)<a href="https://agentmods.dev/commands/hahaxiang27/flowharness/harness.spec-check"><img src="https://agentmods.dev/badge/commands/hahaxiang27/flowharness/harness.spec-check/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/commands/hahaxiang27/flowharness/harness.spec-check"><img src="https://agentmods.dev/badge/commands/hahaxiang27/flowharness/harness.spec-check.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 | $0.00036 | $0.01254 |
| Opus 5 | $0.00018 | $0.00627 |
| Sonnet 5 | $0.00007 | $0.00251 |
| Haiku 4.5 | $0.00004 | $0.00125 |
Grade A, and why
harness-spec-check 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 9d 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 — 114 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Harness Spec 颗粒度审�?
*关联 Constitution 原则 XII · Spec Sizing Discipline�?
**上下文管�?*: 保持当前上下文（需要读 spec.md / tasks.md�?
指令
独立审计某个 feature �?spec 颗粒度，不阻断其他流程。/harness.plan 已经内置同样�?pre-check，本命令是可在任意时刻手工跑的镜像�?
读取以下文件�?1. specs/<feature>/spec.md �?user story 来源
2. specs/<feature>/tasks.md（如存在）�?task 数来�?3. .specify/memory/constitution.md �?取原�?XII 当前阈值（默认 US �?3, Tasks �?30�?
输入参数
- Feature ID：
$ARGUMENTS（必�?· 例如023-learner-loop�?- 找不到对�?specs/<feature>/spec.md�?立即报错并退�?
执行步骤
Step 1 · �?user story
grep -c "^### User Story" specs/<feature>/spec.md
记下数量 N�?
Step 2 · �?task
grep -c "^- \[[ x]\] T[0-9]" specs/<feature>/tasks.md # 如存�?```
记下数量 M。tasks.md 缺失则记�?N/A�?
#### Step 3 · �?acceptance scenario（辅助维度）
```bash
grep -c "^[0-9]\+\. \*\*Given\*\*" specs/<feature>/spec.md
记下数量 K。这个不是硬阈值，�?K 远大�?N 通常意味着 user story 颗粒度太大，每个 story 塞了过多场景�?
Step 4 · 输出审计报告
格式�?
# Spec 颗粒度审�?· <feature> · <YYYY-MM-DD>
## 度量
| 维度 | 实际 | 阈值（原则 XII�?| 状�?|
|---|---|---|---|
| User Stories | N | �?3 | �?/ �?�?X |
| Tasks | M | �?30 | �?/ �?�?Y / �?N/A |
| Acceptance Scenarios | K | 无硬阈�?| 参考�?· K/N = 平均 X �?story |
## 判定
- �?**PASS** · 颗粒度合�?· 可进�?`/harness.plan`
- �?**FAIL** · 必须拆分（见下方建议�?
## 拆分建议（FAIL 时输出）
### 当前 User Stories 列表
1. US1 - <title> (Priority: P1)
2. US2 - <title> (Priority: P2)
...
### 推荐拆分方案
�?user story 边界拆成 N 个独�?spec�?
-1- (US1) -2- (US2) -3- (US3) ...
依次跑：
/speckit.specify "<US1 描述>" /speckit.plan ... /speckit.tasks ... /harness.plan /harness.scope ...
### 拆分依赖图（如有�?spec 依赖�?
提议把共享基础设施（如学员侧整合层 `learner/*`）作为前�?spec，先 ship 完再�?US-level spec�?
## 替代方案
如果你认为不该拆�?*3 �?1**�?
1. **合并 user story** · 仅当几个 US 真正紧密耦合，独立交付不构成完整价�?2. **升级 justification** · �?spec 末尾添加 "## Sizing Override" 段，显式写明不可拆分的根本理�?3. **�?spec**（推�?· 默认�?
未做任一处理 · `/harness.plan` 会一直阻塞在 Step 0 pre-check�?```
### 注意
- **只读命令**：不修改任何 spec 文件，只输出审计报告
- 报告写到 stdout，不写盘（用户可以自己导出）
- Constitution 原则 XII 阈值可调（项目可在自己�?constitution.md 里覆盖默认值），本命令读项目实�?constitution
- 关联文档�? - 原则 XII：`.specify/memory/constitution.md`
- `/harness.plan` Step 0：`.claude/commands/harness.plan.md`
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.
- 9d ago First seen · 114 lines · 0 tokens per session scan A 005bd31c7797
harness-spec-check is a command published in the GitHub repository hahaxiang27/FlowHarness (4 stars, last pushed 2mo ago), licensed MIT. It adds 36 tokens to every session and 1,254 once invoked, about $0.0002 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 commands, from other repositories
template
Manage issue templates for streamlined issue creation.
sync-linear
Sync current work with Linear ticket status.
add-note
Add an internal or external note to a ConnectWise PSA ticket.
fest-show
Show festival progression (in-progress tasks, roadmap, and dependency view).
dispatcher
Pick the next-best repo to work on across the portfolio — rank free repos, recommend one, claim its lease atomically, and route to the entry command.
workpm
A project-management workflow for coordinating multiple AI workers through five stages. It includes task assignment, shared activity logs, worker replacement, and final checks.