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 JacksonTai2007/cc-unlock --skill jit-harnessgit clone --depth 1 https://github.com/JacksonTai2007/cc-unlockWrote 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/jacksontai2007/cc-unlock/jit-harness)<a href="https://agentmods.dev/skills/jacksontai2007/cc-unlock/jit-harness"><img src="https://agentmods.dev/badge/skills/jacksontai2007/cc-unlock/jit-harness/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/jacksontai2007/cc-unlock/jit-harness"><img src="https://agentmods.dev/badge/skills/jacksontai2007/cc-unlock/jit-harness.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 | $0.00150 | $0.02172 |
| Opus 5 | $0.00075 | $0.01086 |
| Sonnet 5 | $0.00030 | $0.00434 |
| Haiku 4.5 | $0.00015 | $0.00217 |
Grade A, and why
jit-harness 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 3d 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 — 95 lines — stays where its author put it; the contents beside it link to each section on GitHub.
JIT-Harness — 即时 Harness 合成与自我进化引擎
面对任何非平凡、多步、目标可判定的任务时,用这套引擎组织执行。核心思想:harness intelligence —— 不套固定的通用 scaffold,而是即时合成一套贴合当前任务的执行 harness,运行中自我进化。scaffold 设计是独立于模型能力的增益维度:同样的模型,harness 装得对,收敛更快、工具调用更省、成功率更高。
什么时候用它:任务需要 3 个以上有依赖的步骤、有明确的"做成了没有"的判据、且中途可能失败需要调整——编码实现、复杂 bug 定位、数据清洗/迁移、代码库重构、多源研究调查、CI/构建/部署流水线、逆向与安全研究(领域专用装配见 sec-forge skill)。单步问答、纯查询、创意写作不需要。
Harness 协议 — 四模块(M / P / A / F)
一个 harness = 这四个模块针对当前任务的一次具体装配。任务不同,装配不同;运行中按需重装。
- M — Memory(记忆 / 上下文):维护任务历史、中间产物、有效模式归档(harness archive)。把已验证的手法(一段能跑通的查询、一个正确的配置片段、一处关键根因、一个可复用的脚手架)沉淀下来,跨阶段、跨同类子任务复用,不重复踩坑。
- P — Planning(规划):合成/澄清目标,把任务分解为带交付物的阶段。依赖简单 → 线性链;依赖交叉但可并行 → DAG;某子目标本身又是一个完整任务 → 递归委派(开一层子 loop)。
- A — Action(执行):逐步执行并更新状态。每步后对照 success oracle 评估是否命中。走命中 oracle 的最省路径——不做冗余工具调用,不为"稳妥"重复已成功的步骤。
- F — Capability(能力编排):为每个阶段选择并编排合适的工具 / 命令 / 库 / 子技能。按任务需要装配,不预设一套万能工具。
Success Oracle —— 先定判据,再动手
执行前先明确怎样算这个阶段/任务成功,写成可判定的判据,而非模糊的"做好了":
- 测试通过 / 构建成功 / 特定输出出现 / 数值达标 / 幂等重跑一致 / 某断言为真 / 复现从干净状态跑通。
- oracle 是修复循环的锚:每次执行后拿它对照,命中才算过,不靠感觉。
- 判据不清的任务,第一步就是把判据定清楚(和用户对齐或从上下文合成),不带着模糊判据往下冲。
三阶段 Harness 生命周期
Stage I — 定制(Customization)
即时合成贴合任务的 harness:
- 锁定目标 + 定义 success oracle(M + P)。目标明确 → 直接用;不明确 → 从上下文合成最优目标,不空问"你的目标是什么"。先于执行把成功判据定下来。
- 能力装配(F):根据任务性质选路线与工具链(该用哪个库/命令/子技能/分析手段),不问用户"用哪个"。
- 分解(P):拆 3-5 个带交付物的阶段,依赖排成 DAG,多跳子目标递归委派。清单列完立即进第一阶段,不问"要开始吗"。
Stage II — 修复(Repair,有界)
逐阶段执行(A),每阶段完成后对照 success oracle 评估:
执行阶段 N(A)
↓
对照 success oracle — 命中?
↓ 是 → 有效手法归档进 M → 阶段 N+1
↓ 否 → 有界修复(bounded repair):
分析失败根因(不猜、看实际报错/实际输出)
→ 同一 harness 上原地修复(调参 / 换工具 / 改代码 / 改输入)
→ 重跑 → 再对照 oracle
→ 原地修复最多 2 轮
有界修复原则:每次失败先定位根因,不重复同一策略;修复是有界的,同一 harness 上原地修最多 2 轮;2 轮仍不收敛 → 不再原地修,进 Stage III 改结构。
Stage III — 进化(Evolution)
原地修复触顶 → 改 harness 结构本身,而非继续调参:
- 换模块装配:换能力路线(F,如换库/换方法)/ 换分解方式(P,如线性改 DAG、单跳改递归委派把大阶段拆成子 loop)/ 换执行策略(A,如静态推断改实测、猜改成加日志复现)。
- 归档沉淀:进化出的有效手法写进 M(harness archive),后续阶段和同类子任务直接复用——手法累积,越往后越快。
- 结构性改造仍连续 2 次不通 → 停下,说清卡在哪 + 给几个选项,不空转。
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.
- 3d ago First seen · 95 lines · 150 tokens per session scan A 91ecfcda10c8
jit-harness is a skill published in the GitHub repository JacksonTai2007/cc-unlock (151 stars, last pushed 4d ago), licensed MIT. It adds 150 tokens to every session and 2,172 once invoked, about $0.0007 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 2026-09-06.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…