Finishing a Development Branch
obra/superpowers
Walks the last step of a branch: confirm tests pass, detect the git environment, ask how to integrate, carry out your choice and clean up the worktree.
将已经可用的 Harness Anything 接入一个尚无台账的目标仓库:检查项目、确认身份、初始化、合并 Agent 入口、建立主控知识位置,并用一个真实任务验证完整流程。缺少源码或 CLI 时先使用 harness-download 完成工具准备,再返回本流程;不重复维护下载步骤。
$ npx skills add FairladyZ625/harness-anything --skill harness-install -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install FairladyZ625/harness-anything harness-install --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/FairladyZ625/harness-anything.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/harness-install .claude/skills/harness-install && 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 "harness-install" agent skill from https://github.com/FairladyZ625/harness-anything/tree/main/skills/harness-install into .claude/skills/harness-install/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "harness-install", 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/FairladyZ625/harness-anything/tree/main/skills/harness-installType 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 FairladyZ625/harness-anything --skill harness-install -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install FairladyZ625/harness-anything harness-install --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/FairladyZ625/harness-anything.git skills-src && mkdir -p .agents/skills && cp -r skills-src/skills/harness-install .agents/skills/harness-install && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "harness-install" agent skill from https://github.com/FairladyZ625/harness-anything/tree/main/skills/harness-install into .agents/skills/harness-install/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "harness-install", 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 FairladyZ625/harness-anything --skill harness-install -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install FairladyZ625/harness-anything harness-install --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/FairladyZ625/harness-anything.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/skills/harness-install .cursor/skills/harness-install && 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 "harness-install" agent skill from https://github.com/FairladyZ625/harness-anything/tree/main/skills/harness-install into .cursor/skills/harness-install/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "harness-install", 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/FairladyZ625/harness-anything.git --path skills/harness-install--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 FairladyZ625/harness-anything --skill harness-install -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install FairladyZ625/harness-anything harness-install --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/FairladyZ625/harness-anything.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/skills/harness-install .gemini/skills/harness-install && 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 "harness-install" agent skill from https://github.com/FairladyZ625/harness-anything/tree/main/skills/harness-install into .gemini/skills/harness-install/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "harness-install", 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 FairladyZ625/harness-anything harness-installInstalls 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 FairladyZ625/harness-anything --skill harness-install -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/FairladyZ625/harness-anything.git skills-src && mkdir -p .github/skills && cp -r skills-src/skills/harness-install .github/skills/harness-install && 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 "harness-install" agent skill from https://github.com/FairladyZ625/harness-anything/tree/main/skills/harness-install into .github/skills/harness-install/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "harness-install", 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 FairladyZ625/harness-anything --skill harness-install -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install FairladyZ625/harness-anything harness-install --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/FairladyZ625/harness-anything.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/skills/harness-install .opencode/skills/harness-install && 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 "harness-install" agent skill from https://github.com/FairladyZ625/harness-anything/tree/main/skills/harness-install into .opencode/skills/harness-install/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "harness-install", 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.
harness-install将已经可用的 Harness Anything 接入一个尚无台账的目标仓库:检查项目、确认身份、初始化、合并 Agent 入口、建立主控知识位置,并用一个真实任务验证完整流程。缺少源码或 CLI 时先使用 harness-download 完成工具准备,再返回本流程;不重复维护下载步骤。
Harness Install is an agent skill from FairladyZ625/harness-anything. 将已经可用的 Harness Anything 接入一个尚无台账的目标仓库:检查项目、确认身份、初始化、合并 Agent 入口、建立主控知识位置,并用一个真实任务验证完整流程。缺少源码或 CLI 时先使用 harness-download 完成工具准备,再返回本流程;不重复维护下载步骤。
Its SKILL.md is about 2.2k tokens, which your agent loads only when the skill is triggered. The skill folder holds 2 other files (for example `agents/openai.yaml`).
It sits in Development. It works with Git. The repository describes itself as: Clean-room Harness rewrite monorepo. The licence is AGPL-3.0.
8 steps, taken from the step headings in SKILL.md.
Read from SKILL.md and the folder at commit c805ddf. It shows what the files ask for, not the result of running them.
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.
Shell commands in SKILL.md call:
gitnodeFrom the folder's file list and the shell code blocks in SKILL.md.
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.
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Harness Install loads about 2.2k tokens when it runs. Until then it costs about 40 tokens; SKILL.md has 413 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.
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.
The full file from FairladyZ625/harness-anything at commit c805ddf, republished under its AGPL-3.0 licence (© FairladyZ625). 413 words, ~2,238 tokens.
.claude/skills/harness-install/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.本技能负责把 Harness 接入具体项目。机器环境、持久源码与用户级技能链接由 harness-download 负责;本技能从可用 CLI 开始。完整引导过程中由同一个 Agent 连续执行,不要求用户在阶段之间重新发指令。
使用本次安装所选源码中的完整技能包,不混用不同版本的技能与 CLI。正式源码分发来自 main,可用 git show origin/main:skills/harness-install/SKILL.md 核对发布分支;明确验证候选分支时使用该候选包并记录版本。
读取项目之前不写入。检查已有 harness/、.harness/、配置及 CLI 可读状态:
init。harness/ 是无关业务目录,或无法判断已有内容:展示真实冲突,先澄清,不覆盖。读取项目布局、Git 状态和历史、现有 AGENTS.md / CLAUDE.md、忽略规则、CI、构建和测试入口。能从磁盘得到的答案不问用户。
| 观察 | 对下一步的影响 |
|---|---|
| 当前初始化方式要求 Git,但项目无仓库或提交 | 在用户授权的目标内先建立必要基础,不操作其他目录 |
| 已有 Agent 指令 | 预备合并草案,保留原指令并取得具体合并批准 |
| 已有忽略规则 | 初始化后回读差异,确认私有台账不会进入公开项目 |
| 有 CI | 记录真实验证入口,首任务按实际结果收口 |
| 没有 CI | 不把完成参数当成真实运行证据;按任务类型取得可验证结果,明确未运行 CI |
| 多人协作 | 提前说明 owner 与评审身份的实际绑定,不能把环境变量当作另一个人 |
| 用户已给首任务 | 直接复用,不再造一个名为“安装测试”的占位任务 |
只合并询问尚缺少的项目标识、首次任务等信息。owner 身份与指令文件合并仍按下文保留具体审批。
复用 harness-download 传来的绝对源码入口和 Node 路径。工具未准备好时读取同一技能包里的下载技能,完成其环境、源码、链接阶段后返回;不在这里再复制 clone 或 npm 安装配方。
下面用 ha 表示本次已核实的 CLI。若使用源码,后续调用均展开为 node "<源码目录>/packages/cli/src/index.ts" ...;临时 shell 函数不会跨工具调用自动保留,不把“node 加参数”装进一个字符串再当命令执行。
检查目标是否已有中心注册和在飞初始化,使用正常的服务根。不要通过 HARNESS_DAEMON_USER_ROOT 为正式安装切换到测试根;若发现现有环境指向隔离根,先核实用途,明确选择正式入口,不能静默把用户已有配置清掉。
读取当前 ha init --help。向用户展示将写入的人员标识、显示名及实际登录身份,取得明确答复;已确认的内容直接复用,不能从 Git 配置猜 owner。
当前纯本地路径也要求 Keycloak 登录及适用的策略组。全新用户在没有身份服务与登录会话的独立 daemon 根中直接执行下一节的 init,会退出 1:
configure_verify_failed
init Configure-Verify smoke failed: Sign in with Keycloak and request an applicable policy group.因此,下载和本地 Unix socket 可用不等于已能初始化项目;不能承诺未登录的新用户直接走通。先按公开服务端运维的首次管理员与登录流程配置身份、登录并取得适用权限,再继续。--person-id 是初始化输入,不能替代认证,也不再使用 unix-socket-owner-boundary 凭据示例。缺少前提时记录该阶段的原始错误及已完成的写入,不删台账重来。
人员与策略组由身份管理入口办理;当前 ha people 只提供执行委托与撤销委托,不提供 add/remove。不手改人员文件,也不用通用文档同步替代身份管理。
命令形状以当前帮助为准,在明确的目标仓库执行:
ha --root "<目标仓库绝对路径>" init --repo-id <项目标识> --person-id <人员标识> --display-name "<显示名>"初次本地安装若要求 Git 基础,先核实是否已有仓库与提交,只补真正缺失的部分,不重置原历史。多个边缘节点不能各自初始化同一个逻辑工作区;由当前有权协调者完成中心写入,其他节点接入已有记录。
回读而不是只看成功回执:
preserved / drifted 中有原 Agent 文件时,确认新入口是否尚未进入,继续下一节。git check-ignore -v harness .harness 和 git ls-files harness .harness 核对私有目录边界,检查实际忽略差异。git check-ignore -q 一次只传一个路径。失败时先识别哪些写入已完成、是否有当前持有者及可用恢复路径。不能自动删台账、清锁或重启服务重来。必要的破坏性清理由用户对具体对象与影响确认;不能影响同一服务下其他项目。
只有原文件存在且 Harness 入口尚未接入时才合并。通过当前 template render 读取仓库模板,如 template://repository/agent-base@1、agent-overlay@1、claude-entry@1,以返回的正文与必需锚点为准,不复制其他人的 AGENTS 文件。
保留原文与项目特有规则;先准备完整合并结果和差异,再让用户批准这一具体修改。保留明确要求的锚点,不用模板覆盖用户内容。同一差异已获批时不再确认。
ha --root "<目标仓库绝对路径>" --json daemon status
ha --root "<目标仓库绝对路径>" --json task list查看目标工作区真实状态和错误,不因文本摘要过短就判断服务损坏。空任务集是新工作区的正常结果。
部分 CLI 回执的 evidence 是 JSON 编码字符串。按实际类型解析,不直接假定字段在顶层;例如读取任务时先解析外层,再解析 evidence。报错与业务数据要分开判断。
随后读取 harness-ceo 的知识初始化方法,建立或复用用户约定、模型矩阵、运行时问题、反馈与执行证据的落点。只建必要索引,不填满空表,不将公共技能中的建议当成用户事实。
复用用户给的首任务,在主控方法下完成真实工作和仓库要求的收口。本阶段与下载引导的首任务是同一个任务。
当前完成契约若要求独立执行评审,应在实施前明确可用评审身份与负责人。可能的拒绝为:
Execution Review requires an independent transport-bound arbiter.HARNESS_ACTOR=agent:... 是身份声明,不是创造独立评审者的方法。不能同一个执行者改名批准自己,也不能清空变量冒充用户的判断。需另一真实执行者、用户判断或宿主支持的独立评审通道时,交付可审阅材料后由该身份完成。只有一个模型不等于只有一个身份,但是否独立必须符合当前契约。
先读 preset-trigger 选择当前预设;一般任务形状如下:
ha --root "<目标仓库绝对路径>" task create --title "<真实任务>" --vertical software/coding --preset standard-task
ha --root "<目标仓库绝对路径>" task start <任务标识>从回执取得真实任务路径,不能手拼派生的 slug。按当前生命周期要求执行,重复启动前先回读已有执行。若启动回执未带执行标识,查询 task show;部分版本将其放在 evidence.lease.executionId,需先解析 evidence 字符串,再读取 ["lease"]["executionId"]。
输出为仓库变更的任务,在本节点第一次 task start 或派工时由 Harness 建立工作树:分支 <任务标识>,目录 .worktrees/<任务标识>,基线为 origin/HEAD 指向的远端默认分支,无远端时取主检出的当前分支。回执写明路径与基线,不需要也不应手动 git worktree add。不改仓库文件的任务没有工作树,task show 的工作区行指向任务包目录。
新工作树需要的准备写在仓库 Settings 的 worktree.setup,按顺序在工作树内只运行一次,环境带 HARNESS_TASK_ID、HARNESS_WORKTREE、HARNESS_REPO_ROOT。init 检测到 npm workspaces 时已写入 node-modules,回执会说明;其他项目按实际需要声明:
ha --root "<目标仓库绝对路径>" settings update --worktree-setup node-modules # Node:镜像根目录 node_modules
ha --root "<目标仓库绝对路径>" settings update --worktree-setup "run: uv sync" # Python,或 "run: pip install -e ."
ha --root "<目标仓库绝对路径>" settings update --worktree-setup "run: make bootstrap" # 单条命令
ha --root "<目标仓库绝对路径>" settings update --worktree-setup none # 清空重复 --worktree-setup 表示多步,顺序即执行顺序;settings read 回读。某步失败时工作树保留、启动或派工被拒,提示写明步骤编号和日志(工作树 Git 目录下 harness-setup/step-<n>.log)。读日志修复原因后重跑同一条 task start 或派工,只重跑尚未成功的步骤;不要删工作树重来。旧命名 codex/<slug>-<id8> 的工作树用 task contract migrate --apply 一次性改名,有未提交改动或在飞执行者的会被列出、不强行移动。
做实际工作并取得证据。交付日志和报告归本次执行;会支撑后续判断的可复核观察按当前事实入口记录,不为凑数量编造事实。仓库规定的事实与关系要求仍然有效。
读取当前决策能力和输入契约;以下展示常见结构,实际字段和合法值以所用版本为准:
{
"title": "<决策标题>",
"question": "<要解决的问题>",
"riskTier": "low",
"urgency": "low",
"decisionClass": "ordinary",
"appliesTo": { "modules": [], "productLines": [] },
"chosen": [{ "id": "CH1", "text": "<选择>", "rationale": "<理由>" }],
"rejected": [{ "id": "RJ1", "text": "<真实替代>", "whyNot": "<未采用原因>" }],
"claims": [], "fulfillments": [], "relations": []
}decisionClass 常用 ordinary 或 standing_policy;不编造替代方案。输入文件放在工作区允许的任务产物位置。遇到 fromFile must stay inside the workspace 时修正输入落点,不能把临时目录当成有效来源。
原始观察先记录,再按当前 ha relation 契约将决策声明关联证据,选择关联派生任务。不要复制退役的 decision relate 命令。用户需要裁决时展示问题、选择、替代和证据后取得批准,由真实有权身份办理;不通过更换环境变量代替用户。一时答不了的问题、上手验收或同意请求,在明确具有关系写权时,用 ha relation relate --type awaits 把任务或决策挂到该人名下(理由写 <question|acceptance|consent|reopen>: <问的是什么>),只有文档或 fact 写权时,把具体问题写进获准报告并请 owner 登记关系。它会出现在对方的 ha agenda「等你处理」和图形界面总览「等你答复」里;对方可在图形界面就地答复,或让 Agent 执行 ha relation unrelate <relation-id> --reason "<答复>",答复后回到提问方的「已答复,待你跟进」。
把收口材料写实,先 doc status 检查候选,再通过受支持的 doc sync --submit 登记本任务文档。不要提交无关候选;路径限定与执行身份的组合以当前能力为准。
收口声明、交付物、验证与风险写入任务包的 closeout.md,通过文档同步接受;task submit 从已接受的文档与本地提交冻结交付,不再接收 JSON 提交文件:
ha --root "<目标仓库绝对路径>" doc sync --submit --task <任务标识>
ha --root "<目标仓库绝对路径>" task submit <任务标识>需要显式指定交付提交时,使用当前帮助中的 --commit <40 位 SHA>;不要用已退役的 --from-file。
代码文档锚通过当前 task code-doc reconcile 在所需阶段产生,不手写机器文件。由真实独立评审者通过 task review-execution 提交评审(参数与文件路径约束先读其 --help);task review 只检查旧评审契约,不能批准完成。评审材料包含:
{ "verdict": "approved", "reason": "<依据>", "evidenceChecked": ["<实际检查的证据>"] }verdict 使用当前合法值,如 approved、changes_requested、dismissed。owner 的同意绑定实际已记录评审与内容版本,不把工作者的报告当作同意。使用当前 task review-consent 和 task complete 契约完成;不添加过期的提交版本参数或自己制造摘要。
task complete 不接受已退役的 --ci passed。CI 证据由完成契约要求的见证来源提供,不能用参数代替运行结果。所选任务无需 CI 时按实际契约处理;无法满足必需门时交付具体缺口,不伪造完成。
最后回读任务是否真的到 done,并检查原始目标确实实现。独立评审尚缺时,准确报告“环境和仓库已就绪,首任务待评审”,不能把整个引导说成已验证。
这是已有中心的接入路线,不能在边缘重新 init 同一个逻辑工作区。先取得中心提供的仓库标识、项目 Git remote、TLS 地址、CA、已注册的节点标识与机器凭据、镜像存放位置及配额;身份与节点注册沿用上面的运维文档。
Clone 目标项目(工具源码下载仍只由 harness-download 维护),确认 Git remote 与中心项目一致。
在 clone 目录注册已有仓库的本地边缘模式:
ha daemon repo register --repo-id <中心仓库标识> --root "<项目 clone 绝对路径>" --mode remote-edge在同一目录做首次同步,参数使用中心给出的真实值:
ha daemon fleet edge sync --host <中心地址> --port <TLS端口> --ca "<CA文件>" --node-id <节点标识> --credential <机器凭据> --view-root "<本地镜像目录>" --quota-bytes <配额字节数>凭据不写入报告或共享 shell 历史;已有受保护的 fleet-edge.json 配置时可省略 --credential,由 CLI 读取。同步要求匹配的已启用 remote-edge 注册,不是把 ledger 当成另一个 Git remote 来拉取。
回读 daemon status 与 task list,确认中心任务在本地可见。后续任务命令使用项目根下的受保护 fleet-edge.json 选择通道;该文件按 packages/daemon/src/client/fleet-edge-config.ts 的 fleet-edge-config/v1 契约准备,不把含机器凭据的文件纳入项目 Git。首次同步成功后仍须确认该配置可用,再按正常任务认领与派工入口工作。注册或同步失败时保留原始错误,不切换为本地初始化。
此入口的命令形状已按 CLI 帮助和路由核对;真实中心、TLS、节点授权与首任务仍须在用户的已授权部署中验收,不能把帮助通过当成 Fleet 已接通。
交付绝对 CLI 入口、目标工作区、用户知识索引、首任务与其证据、真实评审身份及未完成项。工具进入 PATH 的选项由下载技能统一负责,本技能不另提供第二套安装命令。
说明台账与项目各自的历史边界,不在台账里手工补 git commit。列出项目仍未提交的文件,不替用户隐瞒这些变化。下一次使用 harness-ceo,另一个新项目直接使用本技能。
done。© FairladyZ625, AGPL-3.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
SKILL.md and 1 other file in skills/harness-install of FairladyZ625/harness-anything.
Open the folder on GitHubat commit c805ddf
Harness Install 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 |
|---|---|---|---|---|---|---|
| Harness Install this skillFairladyZ625/harness-anything | 226 | — | ~2.2k | Automated safety check: Pass | AGPL-3.0 | |
| Finishing a Development Branchobra/superpowers | 297k | 5 repos | ~1.9k | Automated safety check: Pass | MIT | |
| Code Review ChecklistshareAI-lab/learn-claude-code | 78k | 4 repos | ~1.1k | Automated safety check: Pass | MIT | |
| Code Design Rationale Investigatorcursor/plugins | 11k | 9 repos | ~2.6k | Automated safety check: Pass | None | |
| Contributor-First PR MergeHKUDS/OpenHarness | 16k | 1 repos | ~847 | Automated safety check: Pass | MIT | |
| Finishing A Development Branchfarm-fe/farm | 5.6k | 34 repos | ~1.8k | Automated safety check: Pass | MIT |
obra/superpowers
Walks the last step of a branch: confirm tests pass, detect the git environment, ask how to integrate, carry out your choice and clean up the worktree.
shareAI-lab/learn-claude-code
Reviews code against a five-part checklist covering security, correctness, performance, maintainability and testing, and reports findings in a fixed format.
cursor/plugins
Digs into why code is shaped the way it is by checking git history, pull requests and connected tools in parallel, then reporting a cited read on the tradeoffs.
HKUDS/OpenHarness
Merges external GitHub pull requests while keeping the original author credited, and fixes conflicts after the merge instead of rewriting the contribution.
farm-fe/farm
A skill your agent uses when implementation is complete, all tests pass, and you need to decide how to integrate the work - guides completion of development work by presenting structured options for…
TryGhost/Ghost
Moves a package from another TryGhost repository into Ghost as an internal workspace package while keeping its Git history, with checkpoints for the steps that need an administrator.
FairladyZ625/harness-anything
Contribute a public change to Harness Anything from a GitHub issue through an isolated worktree, scoped tests, manifest-selected gates, a complete bilingual PR body, review triage, and maintainer…
FairladyZ625/harness-anything
在目标仓库 harness.yaml 的 settings.gates 里声明与定制任务完成门:local-command 本地命令见证、manual-attest 人工签字、github-actions CI 见证、none 显式免除,治理字段 mandatorySignoff(双控签注)与 allowOverride(owner 特批),以及…
FairladyZ625/harness-anything
Create, review, or update Harness Anything preset-manifest/v3 packages, including profiles, completion gates, template selections, output shapes, script entrypoints, capability declarations, package…
FairladyZ625/harness-anything
Start Harness Anything task creation by choosing a software/coding preset first.
FairladyZ625/harness-anything
产出与增量维护任务包中的活解释页 artifacts/explainer.html;按物化模板中的唯一权威注释编写,并消费派工注入的 task-context 因果上下文。Use when a task-bound dispatch requires creating, updating, or freezing the task's living explainer page, or…
FairladyZ625/harness-anything
Create, review, or update Harness Anything verticals. An agent skill from FairladyZ625/harness-anything.
Works with
Categories
将已经可用的 Harness Anything 接入一个尚无台账的目标仓库:检查项目、确认身份、初始化、合并 Agent 入口、建立主控知识位置,并用一个真实任务验证完整流程。缺少源码或 CLI 时先使用 harness-download 完成工具准备,再返回本流程;不重复维护下载步骤。. Harness Install is an agent skill from FairladyZ625/harness-anything.
Harness Install fits situations like: development work in your project.
Run `npx skills add FairladyZ625/harness-anything --skill harness-install -a claude-code`. Or copy the skill folder (skills/harness-install in FairladyZ625/harness-anything) into .claude/skills/harness-install in your project. Claude Code loads it when a task matches its description.
Run `npx skills add FairladyZ625/harness-anything --skill harness-install -a codex`. Or copy the skill folder (skills/harness-install in FairladyZ625/harness-anything) into .agents/skills/harness-install in your project. Codex loads it when a task matches its description.
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add FairladyZ625/harness-anything --skill harness-install -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/harness-install, .gemini/skills/harness-install, .github/skills/harness-install and .opencode/skills/harness-install in your project.
Going by SKILL.md and its folder, Harness Install needs the command-line tools its instructions call (git and node). Our summary lists: Python 3.
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.
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.
Harness Install is published under the AGPL-3.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
About 2.2k tokens (SKILL.md is roughly 9k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
Skills that share tags, products or a category with Harness Install: Finishing a Development Branch (obra/superpowers, 297k stars), Code Review Checklist (shareAI-lab/learn-claude-code, 78k stars), Code Design Rationale Investigator (cursor/plugins, 11k stars) and Contributor-First PR Merge (HKUDS/OpenHarness, 16k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
FairladyZ625 (a GitHub user) maintains it in FairladyZ625/harness-anything, which has 226 GitHub stars. The repository holds 12 skills in this directory. The repository was last updated on October 10, 2026.
Source: FairladyZ625/harness-anything on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.