Effective Harnesses
liangdabiao/exa-research-mcp-skill
Long-running agent project harness for Codex, OpenClaw, Claude Code, and other coding agents.
Runs a read-only assessment in one of two modes, a repository health check or a runtime extension gap review, and reports findings as a markdown table.
SKILL.md written in Chinese; this summary is our English description.
Install Repository Health and Gap Assessor
$ npx skills add openJiuwen-ai/agent-core --skill assess -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install openJiuwen-ai/agent-core assess --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/openJiuwen-ai/agent-core.git skills-src && mkdir -p .claude/skills && cp -r skills-src/openjiuwen/rsi/harness_rsi/auto_harness/skills/assess .claude/skills/assess && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "assess" agent skill from https://github.com/openJiuwen-ai/agent-core/tree/develop/openjiuwen/rsi/harness_rsi/auto_harness/skills/assess into .claude/skills/assess/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "assess", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/openJiuwen-ai/agent-core/tree/develop/openjiuwen/rsi/harness_rsi/auto_harness/skills/assessType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add openJiuwen-ai/agent-core --skill assess -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install openJiuwen-ai/agent-core assess --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/openJiuwen-ai/agent-core.git skills-src && mkdir -p .agents/skills && cp -r skills-src/openjiuwen/rsi/harness_rsi/auto_harness/skills/assess .agents/skills/assess && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "assess" agent skill from https://github.com/openJiuwen-ai/agent-core/tree/develop/openjiuwen/rsi/harness_rsi/auto_harness/skills/assess into .agents/skills/assess/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "assess", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add openJiuwen-ai/agent-core --skill assess -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install openJiuwen-ai/agent-core assess --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/openJiuwen-ai/agent-core.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/openjiuwen/rsi/harness_rsi/auto_harness/skills/assess .cursor/skills/assess && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "assess" agent skill from https://github.com/openJiuwen-ai/agent-core/tree/develop/openjiuwen/rsi/harness_rsi/auto_harness/skills/assess into .cursor/skills/assess/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "assess", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/openJiuwen-ai/agent-core.git --path openjiuwen/rsi/harness_rsi/auto_harness/skills/assess--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add openJiuwen-ai/agent-core --skill assess -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install openJiuwen-ai/agent-core assess --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/openJiuwen-ai/agent-core.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/openjiuwen/rsi/harness_rsi/auto_harness/skills/assess .gemini/skills/assess && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "assess" agent skill from https://github.com/openJiuwen-ai/agent-core/tree/develop/openjiuwen/rsi/harness_rsi/auto_harness/skills/assess into .gemini/skills/assess/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "assess", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install openJiuwen-ai/agent-core assessInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add openJiuwen-ai/agent-core --skill assess -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/openJiuwen-ai/agent-core.git skills-src && mkdir -p .github/skills && cp -r skills-src/openjiuwen/rsi/harness_rsi/auto_harness/skills/assess .github/skills/assess && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "assess" agent skill from https://github.com/openJiuwen-ai/agent-core/tree/develop/openjiuwen/rsi/harness_rsi/auto_harness/skills/assess into .github/skills/assess/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "assess", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add openJiuwen-ai/agent-core --skill assess -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install openJiuwen-ai/agent-core assess --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/openJiuwen-ai/agent-core.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/openjiuwen/rsi/harness_rsi/auto_harness/skills/assess .opencode/skills/assess && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "assess" agent skill from https://github.com/openJiuwen-ai/agent-core/tree/develop/openjiuwen/rsi/harness_rsi/auto_harness/skills/assess into .opencode/skills/assess/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "assess", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
assess- GitHub stars
- 446
- Token cost
- ~613 tokens
- SKILL.md length
- 172 words
- Files
- 1
- Skills in repo
- 10
- Repo updated
- First seen
- Licence
- Apache-2.0
At a glance
Runs a read-only assessment in one of two modes, a repository health check or a runtime extension gap review, and reports findings as a markdown table.
- Works in 5 steps: 构建状态: 运行 lint/type-check，记录错误数量 → 测试覆盖: 运行测试套件，记录通过率 → 近期变更: git log --oneline -20，识别活跃区域 → …
- Assessing the health of a codebase before planning improvements to it
- SKILL.md covers 模式选择, 代码库健康评估, Runtime Extension 能力缺口评估 and 约束
- Calls git
What it does
Two assessment modes are selected from the query. The mode marker runtime_extension_gap_assessment chooses the extension gap review, while repository_health_assessment, or no marker at all, chooses the repository health check. The skill is written in Chinese for an auto-harness pipeline stage, and the two modes are never mixed.
The health check applies to improving the harness itself, its CLI, pipeline or code quality. It records lint and type-check error counts, the test pass rate, the last 20 commits to spot active areas, module dependencies, file sizes and complexity hotspots, and TODO, FIXME and HACK markers, then compares features, code quality, coverage and documentation in a table with gap, impact, feasibility, a suggested fix and target files.
The extension review targets runtime extensions such as office automation, slide or report generation and external integrations. It looks at the user's goal, the target artifact, inputs and outputs, how work maps onto tools, skills and rails, the missing capabilities and candidate file paths. It does not research other coding agents unless asked, never modifies files, checks past experience through experience_search, and keeps the report between 2000 and 5000 characters.
When your agent uses it
- Assessing the health of a codebase before planning improvements to it
- Finding capability gaps before building a runtime extension such as slide generation
- Producing a gap table with impact, feasibility and target files
Example prompts
- “Run a repository health assessment and list lint errors, test results and TODO hot spots.”
- “Assess capability gaps for a runtime extension that generates PPT reports from templates.”
Requirements
- A repository where lint, type-check and test commands can run
Workflow steps
5 steps, taken from the first numbered list in SKILL.md.
- 构建状态: 运行 lint/type-check，记录错误数量
- 测试覆盖: 运行测试套件，记录通过率
- 近期变更: git log --oneline -20，识别活跃区域
- 架构健康: 检查模块依赖、文件大小、复杂度热点
- 已知问题: 搜索 TODO/FIXME/HACK 标记
What it can do on your machine
Read from SKILL.md and the folder at commit 9824919. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
Shell commands in SKILL.md call:
git
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md. Its commands use git, which can reach the network depending on how they are called.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
Repository Health and Gap Assessor loads about 613 tokens when it runs. Until then it costs about 16 tokens; SKILL.md has 172 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
SKILL.md
The full file from openJiuwen-ai/agent-core at commit 9824919, republished under its Apache-2.0 licence (© openJiuwen-ai). 172 words, ~613 tokens.
Download SKILL.mdSave it as
.claude/skills/assess/SKILL.md (or your agent's skills folder).- name
- assess
- description
- 评估方法论 — 根据 query 中的评估模式执行代码库健康评估或 runtime extension 能力缺口评估
- immutable
- true
- tools
- read_file, glob_tool, grep_tool, list_dir, experience_search
Assess Skill
你是评估阶段的 agent，负责根据 query 中的 评估模式 选择评估方法。
模式选择
先读取 query 中的模式标记：
评估模式: runtime_extension_gap_assessment：执行 Runtime Extension 能力缺口评估。评估模式: repository_health_assessment或未提供模式：执行代码库健康评估。
不要把两种模式混在一起。Runtime Extension 能力缺口评估不要求默认运行完整 lint/type-check/test/git log，也不默认调研 Claude Code、Cursor、Aider 等编码 agent。
代码库健康评估
适用于优化 auto-harness 自身、增强 CLI、修 pipeline、改代码质量或改善开发体验。
必读清单
执行评估前，必须检查以下内容：
- 构建状态: 运行 lint/type-check，记录错误数量
- 测试覆盖: 运行测试套件，记录通过率
- 近期变更:
git log --oneline -20，识别活跃区域 - 架构健康: 检查模块依赖、文件大小、复杂度热点
- 已知问题: 搜索 TODO/FIXME/HACK 标记
能力对比方法
对比维度：
- 功能完整性（0-1 分）
- 代码质量（lint 错误密度）
- 测试覆盖率
- 文档完整性
- 性能指标（如有）
输出格式：markdown 表格，列：
竞品 | 功能 | 当前状态 | 差距描述 | 影响 | 可行性 | 建议方案 | 目标文件
输出结构
评估报告必须包含：
# 评估报告
## 构建状态
- lint 错误数: N
- type-check 错误数: N
## 测试状态
- 通过: N / 总计: N
- 失败用例列表
## 架构观察
- 模块依赖关系
- 复杂度热点（>200 行的文件）
## 改进方向
- 按优先级排序的建议列表Runtime Extension 能力缺口评估
适用于创建或优化运行时扩展，例如办公自动化、PPT/报告生成、领域知识注入、外部系统集成、文件处理、上下文增强等。
调研重点
- 用户目标：用户要自动化什么工作流，目标用户是谁，成功结果是什么。
- 目标产物：产物类型、文件格式、结构要求、质量要求、品牌/领域约束。例如 PPT 生成要关注模板、版式、企业风格、图片/表格/图表、可编辑性。
- 输入输出：输入来自自然语言、文件、模板、知识库、仓库内容还是外部 API；输出需要落盘、返回消息、调用工具还是注册为 runtime extension。
- 组件映射：Tool 用于实际执行动作；Skill 用于承载领域知识、生成规范、流程指导；Rail 只在需要生命周期拦截、上下文增强或后台行为时使用。
- 能力缺口：识别缺少的工具实现、领域 skill、模板资源、文件生成库、配置加载、验收样例和验证方式。
- 目标文件：给出后续 design/implement 阶段可使用的候选文件路径，优先指向 runtime extension 的 tools、rails、skills、配置和测试位置。
竞品/参考对象规则
不要默认研究 Claude Code、Cursor、Aider 或主流编码 agent。只有用户明确要求参考某个竞品、工具、产品或开源项目时，才围绕该对象调研。
如果用户没有指定竞品，缺口来源应来自用户需求和领域范式，例如：
用户需求办公自动化PPT生成工具报告生成流程领域范式
输出格式
输出 markdown 表格，列必须为：
竞品 | 功能 | 当前状态 | 差距描述 | 影响(0-1) | 可行性(0-1) | 建议方案 | 目标文件
为兼容解析器，保留“竞品”列；但在本模式下该列表示来源/参考对象，不要求是真实竞品。
约束
- 只读: 不得修改任何文件
- 代码库健康评估的结论和后续建议必须遵守本轮变更范围：
源码只允许
openjiuwen/harness/**、openjiuwen/core/**； 这两个源码目录下的模块内README.md/ Markdown 也允许修改，例如openjiuwen/harness/cli/README.md； 配套文件只允许tests/**、examples/**； 仓库级文档只允许docs/en/、docs/zh/下的 Markdown 文件 - Runtime Extension 能力缺口评估必须以 query 中的用户目标和 pipeline 约束为准；如果目标明确涉及 auto-harness 或 runtime extension 目录，可以把相关路径作为候选目标文件
- 使用
experience_search查询历史经验，避免重复评估 - 评估报告长度控制在 2000-5000 字符
© openJiuwen-ai, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
Just SKILL.md in openjiuwen/rsi/harness_rsi/auto_harness/skills/assess of openJiuwen-ai/agent-core.
Open the folder on GitHubat commit 9824919
Compare with similar skills
Repository Health and Gap Assessor next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Repository Health and Gap Assessor this skillopenJiuwen-ai/agent-core | 446 | — | ~613 | Automated safety check: Pass | Apache-2.0 | |
| Effective Harnessesliangdabiao/exa-research-mcp-skill | 110 | — | ~1.5k | Automated safety check: Pass | None | |
| Code Reviewaiskillstore/marketplace | 433 | — | ~2.3k | Automated safety check: Pass | None | |
| Qt C++ Code Reviewx-tools-author/x-tools | 1.1k | 2 repos | ~4.3k | Automated safety check: Pass | BSD-3-Clause | |
| Skill Doctorwarpdotdev/common-skills | 610 | 2 repos | ~2.6k | Automated safety check: Pass | MIT | |
| Ponytail Debt LedgerDietrichGebert/ponytail | 160k | — | ~453 | Automated safety check: Pass | MIT |
Similar skills
Code Review
aiskillstore/marketplace
This skill should be used when the user requests a code review of changed files.
Qt C++ Code Review
x-tools-author/x-tools
Read-only review of Qt6 C++ code that combines a deterministic lint script with six parallel analysis agents and reports only high-confidence issues.
Skill Doctor
warpdotdev/common-skills
Grades agent skills by scoring agent conversations for efficiency, code quality, procedure compliance, and verbosity, then drafts concrete skill edits and a shareable report.
Ponytail Debt Ledger
DietrichGebert/ponytail
Collects every ponytail: comment in a codebase into one debt ledger, flags shortcuts with no upgrade trigger and reports without changing any files.
Worktrunk CLI Output Rules
max-sixty/worktrunk
CLI output standards for worktrunk: message functions, ANSI color nesting and the shell integration that changes directory after the wt command exits.
More from openJiuwen-ai/agent-core
All 10 skills in this repoSkill Safety Evaluator
openJiuwen-ai/agent-core
Static safety audit of a SKILL.md that scores five dimensions and acts as a gate: skills below the pass line do not ship, whatever else they score.
Engineering Communication Rules
openJiuwen-ai/agent-core
Chinese-language rules for how an agent writes commit messages, PR descriptions, session journals, handoff issues and requests for help.
Python Patterns for agent-core
openJiuwen-ai/agent-core
Reference for idiomatic Python in the agent-core codebase: immutability, protocols, exception hierarchies, context managers and async patterns.
Agent-Core Python Testing
openJiuwen-ai/agent-core
Pytest patterns for the agent-core codebase: a red-green-refactor workflow, conftest fixtures, custom marks, monkeypatch and patch mocking, and async tests.
Agent-Core Security Checklist
openJiuwen-ai/agent-core
A ten-category security checklist for the agent-core codebase, to run before any security-sensitive change or pull request: secrets, input validation, SQL, access control and prompt injection.
Verification Loop
openJiuwen-ai/agent-core
Formalizes agent-core's make check/type-check/test/fix pipeline into a structured 6-phase verification skill.
Related
Works with
Categories
Questions about Repository Health and Gap Assessor
What does Repository Health and Gap Assessor do?
Runs a read-only assessment in one of two modes, a repository health check or a runtime extension gap review, and reports findings as a markdown table. Two assessment modes are selected from the query. The mode marker runtime_extension_gap_assessment chooses the extension gap review, while repository_health_assessment, or no marker at all, chooses the repository health check.
When should I use Repository Health and Gap Assessor?
Repository Health and Gap Assessor fits situations like: assessing the health of a codebase before planning improvements to it; finding capability gaps before building a runtime extension such as slide generation; producing a gap table with impact, feasibility and target files.
How do I install Repository Health and Gap Assessor in Claude Code?
Run `npx skills add openJiuwen-ai/agent-core --skill assess -a claude-code`. Or copy the skill folder (openjiuwen/rsi/harness_rsi/auto_harness/skills/assess in openJiuwen-ai/agent-core) into .claude/skills/assess in your project. Claude Code loads it when a task matches its description.
How do I install Repository Health and Gap Assessor in Codex?
Run `npx skills add openJiuwen-ai/agent-core --skill assess -a codex`. Or copy the skill folder (openjiuwen/rsi/harness_rsi/auto_harness/skills/assess in openJiuwen-ai/agent-core) into .agents/skills/assess in your project. Codex loads it when a task matches its description.
Can I use Repository Health and Gap Assessor in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add openJiuwen-ai/agent-core --skill assess -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/assess, .gemini/skills/assess, .github/skills/assess and .opencode/skills/assess in your project.
What does Repository Health and Gap Assessor need to run?
Going by SKILL.md and its folder, Repository Health and Gap Assessor needs the command-line tools its instructions call (git). Our summary lists: A repository where lint, type-check and test commands can run.
Does Repository Health and Gap Assessor access the network?
SKILL.md contains no URLs. Its commands use git, which can reach the network depending on how they are called. This is read from the text; nothing was executed.
Is Repository Health and Gap Assessor safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
What licence does Repository Health and Gap Assessor use?
Repository Health and Gap Assessor is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Repository Health and Gap Assessor use?
About 613 tokens (SKILL.md is roughly 2.5k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
What are the alternatives to Repository Health and Gap Assessor?
Skills that share tags, products or a category with Repository Health and Gap Assessor: Effective Harnesses (liangdabiao/exa-research-mcp-skill, 110 stars), Code Review (aiskillstore/marketplace, 433 stars), Qt C++ Code Review (x-tools-author/x-tools, 1.1k stars) and Skill Doctor (warpdotdev/common-skills, 610 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Repository Health and Gap Assessor?
openJiuwen-ai (a GitHub organization) maintains it in openJiuwen-ai/agent-core, which has 446 GitHub stars. The repository holds 10 skills in this directory. The repository was last updated on October 11, 2026.
Source: openJiuwen-ai/agent-core on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.