ccc Semantic Code Search
cocoindex-io/cocoindex-code
Semantic code search and index management with the ccc CLI: the agent initializes, indexes and queries the project by concept, filtering by language or path.
Coder-agent working-file budget discipline: keep the editable working set (files you /add into writable context) under ~25k tokens, separate "read" from "edit", delegate breadth to a read-only…
$ npx skills add agentsope/SkillAlchemy --skill agentsop-context-scope-discipline -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install agentsope/SkillAlchemy agentsop-context-scope-discipline --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/agentsope/SkillAlchemy.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/agentsop-context-scope-discipline .claude/skills/agentsop-context-scope-discipline && 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 "agentsop-context-scope-discipline" agent skill from https://github.com/agentsope/SkillAlchemy/tree/master/skills/agentsop-context-scope-discipline into .claude/skills/agentsop-context-scope-discipline/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "agentsop-context-scope-discipline", 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/agentsope/SkillAlchemy/tree/master/skills/agentsop-context-scope-disciplineType 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 agentsope/SkillAlchemy --skill agentsop-context-scope-discipline -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install agentsope/SkillAlchemy agentsop-context-scope-discipline --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/agentsope/SkillAlchemy.git skills-src && mkdir -p .agents/skills && cp -r skills-src/skills/agentsop-context-scope-discipline .agents/skills/agentsop-context-scope-discipline && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "agentsop-context-scope-discipline" agent skill from https://github.com/agentsope/SkillAlchemy/tree/master/skills/agentsop-context-scope-discipline into .agents/skills/agentsop-context-scope-discipline/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "agentsop-context-scope-discipline", 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 agentsope/SkillAlchemy --skill agentsop-context-scope-discipline -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install agentsope/SkillAlchemy agentsop-context-scope-discipline --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/agentsope/SkillAlchemy.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/skills/agentsop-context-scope-discipline .cursor/skills/agentsop-context-scope-discipline && 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 "agentsop-context-scope-discipline" agent skill from https://github.com/agentsope/SkillAlchemy/tree/master/skills/agentsop-context-scope-discipline into .cursor/skills/agentsop-context-scope-discipline/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "agentsop-context-scope-discipline", 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/agentsope/SkillAlchemy.git --path skills/agentsop-context-scope-discipline--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 agentsope/SkillAlchemy --skill agentsop-context-scope-discipline -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install agentsope/SkillAlchemy agentsop-context-scope-discipline --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/agentsope/SkillAlchemy.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/skills/agentsop-context-scope-discipline .gemini/skills/agentsop-context-scope-discipline && 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 "agentsop-context-scope-discipline" agent skill from https://github.com/agentsope/SkillAlchemy/tree/master/skills/agentsop-context-scope-discipline into .gemini/skills/agentsop-context-scope-discipline/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "agentsop-context-scope-discipline", 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 agentsope/SkillAlchemy agentsop-context-scope-disciplineInstalls 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 agentsope/SkillAlchemy --skill agentsop-context-scope-discipline -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/agentsope/SkillAlchemy.git skills-src && mkdir -p .github/skills && cp -r skills-src/skills/agentsop-context-scope-discipline .github/skills/agentsop-context-scope-discipline && 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 "agentsop-context-scope-discipline" agent skill from https://github.com/agentsope/SkillAlchemy/tree/master/skills/agentsop-context-scope-discipline into .github/skills/agentsop-context-scope-discipline/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "agentsop-context-scope-discipline", 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 agentsope/SkillAlchemy --skill agentsop-context-scope-discipline -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install agentsope/SkillAlchemy agentsop-context-scope-discipline --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/agentsope/SkillAlchemy.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/skills/agentsop-context-scope-discipline .opencode/skills/agentsop-context-scope-discipline && 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 "agentsop-context-scope-discipline" agent skill from https://github.com/agentsope/SkillAlchemy/tree/master/skills/agentsop-context-scope-discipline into .opencode/skills/agentsop-context-scope-discipline/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "agentsop-context-scope-discipline", 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.
agentsop-context-scope-disciplineCoder-agent working-file budget discipline: keep the editable working set (files you /add into writable context) under ~25k tokens, separate "read" from "edit", delegate breadth to a read-only…
Agentsop Context Scope Discipline is an agent skill from agentsope/SkillAlchemy. Coder-agent working-file budget discipline: keep the editable working set (files you /add into writable context) under ~25k tokens, separate "read" from "edit", delegate breadth to a read-only repo-map, and drop files once edited. Use when an LLM coder-agent edits multiple files, when the working set must stay focused, or when the model starts editing the wrong file / missing targets because too much context dilutes attention. Search keywords: context window full, agent edits wrong file, too much context, /add…
Its SKILL.md is about 3k tokens, which your agent loads only when the skill is triggered. The skill folder holds 5 other files, including reference files (for example `README.md`, `intermediate/operation_candidates.json` and `references/R1-source-evidence.md`).
It sits in Agent Workflows, covering Codebase onboarding and Context engineering. The repository describes itself as: From thought to skill. From signal to structure. The licence is MIT.
7 steps, taken from the step headings in SKILL.md.
Read from SKILL.md and the folder at commit d0f0355. 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.
No scripts in the folder and no shell commands in SKILL.md.
From the folder's file list and the shell code blocks in SKILL.md.
No URLs in SKILL.md.
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.
Agentsop Context Scope Discipline loads about 3k tokens when it runs, and up to ~4.3k if it reads all its reference files. Until then it costs about 155 tokens; SKILL.md has 789 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 agentsope/SkillAlchemy at commit d0f0355, republished under its MIT licence (© agentsope). 789 words, ~3,014 tokens.
.claude/skills/agentsop-context-scope-discipline/SKILL.md (or your agent's skills folder). This skill also uses 3 other files; get the full folder from GitHub.一句话:编辑代码时,工作文件预算(你
/add-ed 进可写上下文的文件)要压在 ~25k tokens 以内。超过这个量,"more context ≠ better edits"——模型注意力被稀释,开始改错文件、漏看你刚加进去的目标。广度交给 [[agentsop-repo-map]](只读签名地图),深度只留给"这次真要编辑"的那几个文件。
这是一个增强叠加技能(enhance overlay)。它不替代任何"通用 token 预算"建议,而是把那条泛泛的"少塞上下文"打磨成一条 coder-agent 专属的硬规则:区分"读"与"改",只把"改"的文件加进工作集。借用 Aider 的实测阈值——
"Above about 25k tokens of context, most models start to become distracted." [aider.chat/docs/troubleshooting/edit-errors.html]
下列任一情形成立时,把"工作文件预算纪律"作为该编辑会话的标准约束:
/tokens(或等价物)逼近 25k;响应被截断;长会话里模型"记住了错的东西"。不应激活的反面信号:单文件已知的小改动(工作集天然就是 1);纯讨论/架构问答(用只读上下文 + [[agentsop-repo-map]] 即可,不进工作集);非编辑任务。
more context ≠ better edits. 过了约 ~25k tokens 的文件量,模型就开始失焦——只把你这一轮真要编辑的文件加进工作集,其余的靠 [[agentsop-repo-map]] 顶上。
LLM 看到的编辑上下文分三层,优先级与写权限递减:
| 层 | 内容 | 写权限 | 预算策略 |
|---|---|---|---|
| 系统提示 + 编辑格式 | harness 固化 | harness | 不可控 |
| 只读上下文 | [[agentsop-repo-map]] 签名地图 + /read 的参考文件 + CONVENTIONS.md | 人/agent 配置 | 给"广度"——用地图覆盖全仓,但只放签名不放函数体 |
| 工作集(写集合) | /add-ed 的文件 | LLM 唯一能编辑的 | 给"深度"——只放这次真要改的,压在 ~25k 以内 |
核心区分:repo-map 给"哪儿"(breadth,签名级,便宜),工作集给"怎么改"(depth,全文级,贵)。把这两种需求混进同一个篮子("全
/add进来再说")是本技能要根除的反模式。
25k 不是"塞到 25k 就崩",而是"过了 25k 编辑准确率开始断崖式下降"。它是个信号阈:
凭直觉,"我要理解这 10 个文件才能改对,那就全 /add"。实测相反:
Aider 只靠 repo-map(不把文件加进工作集)在 SWE-Bench Lite 上仍 70.3% 命中正确文件 [aider.chat/2024/05/22/swe-bench-lite.html]。
即"找文件"这件事不需要把文件灌进工作集——只读地图就够了。工作集只为"编辑"存在。把这两件事拆开,是省预算的关键。详见 [[agentsop-repo-map]]。
预算是一份蛋糕,不是各自独立的盘子。/add 了正确文件后,[[agentsop-repo-map]] 应自动缩小("adjusts ... based on the state of the chat" [aider.chat/docs/repomap.html]),把 token 让给真代码。如果你的 harness 不会自动缩地图,编辑期就手动 --map-tokens 调小或归零。
本技能管文件维度(工作集里有哪些文件);[[agentsop-session-state-hygiene]] 管历史维度(对话历史是否污染当前任务)。二者共用同一份 25k 预算:
/drop 不再需要的文件(本技能);/clear 清历史([[agentsop-session-state-hygiene]])。/drop 保历史去文件;/clear 保文件去历史;/reset 两者都丢。任务进来,第一步不是 /add,而是分类。对每个相关文件问一句:"这一轮我会修改它的字节吗?"
会改它的字节 → 候选写集合(稍后 /add)
只需理解它的契约 → 只读:/read,或干脆只靠 repo-map 的签名
不确定改哪些 → 先不加任何文件,进 Phase 2 让 repo-map 帮你定位经验法则:写集合目标 ≤ 5 个文件。超过,多半是任务没拆够。
> /ask which files implement <feature>?
< [模型基于只读 repo-map 回答候选文件]模型命名出目标后,你再决定把哪些加进工作集(Op locate-then-add)。"找文件"和"改文件"永远两步走——这是 [[agentsop-repo-map]] 与本技能共享的设计哲学。
/add 你会编辑的,参考文件用 /read/add src/auth.py tests/test_auth.py # 这两个会改 → 进写集合
/read src/config.py docs/auth.md # 只参考,不改 → 只读铁律重申:少 /add,敢 /drop。"为了保险全加"恰恰是让模型改错文件的主因。
/tokens # 看当前占用;接近 25k 是黄灯| 信号 | 动作 |
|---|---|
/tokens 逼近 25k | /drop 已经改完、不再相关的文件 |
| repo-map 占比偏大 | 调小 --map-tokens(目标文件已定,地图可缩) |
| 模型反复改错文件 | /ls 检查工作集;/drop 多余的,/add 缺的 |
| 历史漂移(不是文件问题) | 转交 [[agentsop-session-state-hygiene]]:/clear |
/drop 它工作集不是"会话期一直累积"的。某文件这一轮的修改告一段落、后续子任务不再碰它——立即 /drop。把腾出的预算还给下一批要改的文件。这是把工作集当滑动窗口用,而不是当垃圾堆。
地图也缩了、能 /drop 的都 /drop 了,预算还是破 25k?这是任务太宽的信号,不是预算的问题:
1. 进子目录 + --subtree-only(缩小 repo-map 范围,见 [[agentsop-repo-map]] §3)
2. 拆任务:大需求拆成多个收敛子目标,每个子目标一个会话
3. 每个新会话只带它真正要改的那 ≤5 个文件每条给 Trigger / Action / Output / Evidence。命令名以 Aider 为参考,行为框架无关。
classify(file) 区分"读"与"改"/read 或仅 repo-map 签名)。locate-then-add(task) 先定位再加/add 命名出的目标。add(files) 加入工作集(克制)/add 只加该文件。不加"可能会用到"的、不加整目录。/add 或你 /add 了太多无关文件" [aider.chat/docs/usage].read(files) 加为只读参考/read,进只读层,LLM 不能编辑。--read / /read;只读层与读写层分离 [aider.chat/docs/usage/conventions.html]。budget_watch() 监控预算/tokens 看占用分布(工作集 / 历史 / 地图各占多少)。drop(files) 改完即释放/drop 该文件,腾出预算。/drop 是常态操作而非应急;动态预算 [aider.chat/docs/repomap.html]。lean-on-map() 把广度还给地图/drop + 信任 repo-map)。必要时 /map 审计地图已覆盖什么。split-task() 拆任务而非塞预算触发:一个改动横跨 10 个文件的调用链,但你实际只会修改 2 个(比如改一个 API 签名 + 它的一处实现)。本能是把 10 个全 /add 进来"看全"。
诊断:你把"理解广度"误当成"编辑深度"。10 个里有 8 个你只需要看签名/契约,不会动它们的字节。
决策规则:
| 文件角色 | 数量 | 放哪 |
|---|---|---|
| 真要改字节 | 2 | /add(写集合) |
| 需看完整契约/会被这次改动影响、要核对 | 1–2 | /read(只读全文) |
| 只需知道"它在哪、签名是什么" | 6–7 | 不加,靠 [[agentsop-repo-map]] 签名 |
为什么有效:repo-map 只读即 70.3% 命中正确文件 [aider.chat/2024/05/22/swe-bench-lite.html]——广度不需要进工作集。把 2 个进写集合、地图覆盖其余 8 个,预算从"10 个全文"降到"2 全文 + 8 签名",编辑注意力集中在真正要改的两处。
反模式:10 个全 /add → 破 25k → 模型在 8 个无关文件里挑了错的位置改 → 回滚 → 重来。
触发:多文件重构进行中,/tokens 显示已过 25k,模型开始截断响应、漏看你刚加的文件。
决策树(按代价递增,能停就停):
| 步 | 动作 | 何时停 |
|---|---|---|
| 1 | /tokens 看占用分布:工作集 / 历史 / 地图谁是大头 | 找到主要占用者 |
| 2 | 已改完的文件 /drop(Op 6) | 工作集回到 ≤5、预算降到 25k 下 |
| 3 | 地图占比大 → --map-tokens 调小或归零(目标已定,地图可让位) | 预算回落 |
| 4 | 历史是大头、且话题已切 → 转 [[agentsop-session-state-hygiene]]:/clear(保文件去历史) | 历史清掉 |
| 5 | 仍破 25k → 任务太宽:split-task(),余下子任务新开会话 | 单会话扛得住 |
关键认知:先动文件维度(drop / 缩地图),再动历史维度(clear),最后才拆任务。/drop 与 /clear 是互补而非二选一——前者是本技能,后者是 [[agentsop-session-state-hygiene]]。
反模式:预算满了第一反应是"换更大上下文窗口的模型"。窗口更大不改变 25k 稀释阈——大窗口模型塞到 25k+ 一样失焦。先收工作集,别先换模型。
触发:连续几轮,模型的 diff 落在错误文件上。直觉是"它没看够,再 /add 几个"。
诊断:方向反了。改错文件的两种根因都不靠"加更多文件"解决:
| 现象 | 根因 | 修复 |
|---|---|---|
| 改的文件根本没在工作集里 | 目标没 /add | /add 目标文件(Op 3) |
| 工作集里文件太多、它挑错了 | /add 过量稀释 | /drop 无关的,收到 ≤5(Op 6) |
决策规则:/ls 看现状 → 缺目标就加目标、多余就 drop → 不知道哪个是目标就 locate-then-add(Op 2)让地图替你找。几乎不会是"加更多文件"能解的。
/add 整个目录——"这个 feature 在 src/payments/,全加"。目录里 90% 的文件你不会改,纯稀释。只 /add 那 2–3 个目标文件。/drop——把工作集当只进不出的垃圾堆。改完的文件不释放,预算单调上涨直到破 25k。/drop 是常态操作。/add 进可写上下文。读懂用 /read 或 repo-map 签名,编辑才用 /add。/drop 与 /clear——文件多就 /drop(本技能);历史脏就 /clear([[agentsop-session-state-hygiene]])。用错维度解决不了问题。同一条"只把要改的文件放进工作集"纪律,四个 harness 各自的接口与默认行为:
| Aider | Claude Code | Cursor | Cline | |
|---|---|---|---|---|
| 加入工作集 | /add <files>(显式,仅这些可编辑) | Read 工具按需读文件入上下文 | @file / @folder mention | 按需 read_file tool-call |
| 移出/释放 | /drop <files> | 上下文压缩 / /clear | 移除 mention | tool-call 历史自然滚出 |
| 只读参考 | /read <file>(不可编辑) | 读了即在上下文(无读/写区分) | @file 同样方式 | 同上,无显式只读层 |
| 广度来源 | tree-sitter repo-map(签名) | Grep/Glob/Read 按需探索 | 全仓 codebase index | 文件树 + 主动读 |
| 预算监控 | /tokens(25k 显式建议) | 上下文窗口指示 + 自动压缩 | 闭源 | 上下文长度可见 |
| 写权限边界 | 硬:仅 /add 的可编辑 | 软:能读即能改(用工具白名单约束) | 软:可改任意打开文件 | 软:可改任意读过的文件 |
关键差异:
/read vs /add),最贴合本技能——工作集就是写白名单。其 25k 阈值是这条纪律的实测来源。Read 进来的文件既可读也可被 Edit。本技能在这里表现为自律——不要为了"看全"而 Read 整个目录;用 Grep/Glob 定位(相当于 repo-map 的广度),只 Read 你要 Edit 的文件。/clear 与自动压缩对应 /drop 的预算回收。@-mention 选上下文;mention 越多预算越紧。纪律是"@ 你要改的,别 @ 整个 folder 求保险"。统一心智:无论接口是 /add、@file、Read 还是 read_file,规则不变——进工作集的应当是"这一轮会编辑的文件",广度交给地图/搜索,预算盯住 ~25k。
主要:
/add /read /drop /tokens 命令面。派生:
references/R1-source-evidence.md — 来源逐条 quote。intermediate/operation_candidates.json — 操作抽取过程。关联技能:
/drop(文件)与 /clear(历史)共享同一份 25k 预算。跨工具(一般认知):Aider / Claude Code / Cursor / Cline 公开文档与博客。
© agentsope, MIT. 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 3 other files (references) in skills/agentsop-context-scope-discipline of agentsope/SkillAlchemy.
Open the folder on GitHubat commit d0f0355
Agentsop Context Scope Discipline 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 |
|---|---|---|---|---|---|---|
| Agentsop Context Scope Discipline this skillagentsope/SkillAlchemy | 466 | — | ~3k | Automated safety check: Pass | MIT | |
| ccc Semantic Code Searchcocoindex-io/cocoindex-code | 2.8k | — | ~938 | Automated safety check: Pass | Apache-2.0 | |
| Repomix Codebase Packeryamadashy/repomix | 29k | — | ~1.3k | Automated safety check: Notes | MIT | |
| Codebase Exploreryologdev/yoyo-evolve | 1.9k | — | ~2.8k | Automated safety check: Pass | MIT | |
| Caveman Repository ExplorerJuliusBrussee/caveman | 111k | 1 repos | ~492 | Automated safety check: Pass | Apache-2.0 | |
| Ripwire Codebase Orientationredhat-et/ripwire | 2.4k | — | ~4.5k | Automated safety check: Notes | Apache-2.0 |
cocoindex-io/cocoindex-code
Semantic code search and index management with the ccc CLI: the agent initializes, indexes and queries the project by concept, filtering by language or path.
yamadashy/repomix
Packs a local directory or remote GitHub repository into one AI-friendly file with Repomix, then searches it to explore structure, find patterns and count tokens.
yologdev/yoyo-evolve
Builds a structural map of a large or unfamiliar codebase by dispatching sub-agents to summarize regions, keeping the main context small.
JuliusBrussee/caveman
A read-only explorer for cold-start orientation or failed searches that replies with nothing but file path and line range citations, keeping its reads out of main context.
redhat-et/ripwire
Maps an unfamiliar repo or subsystem with the ripwire CLI before reading files, climbing an escalation ladder only until you are oriented.
proffesor-for-testing/agentic-qe
Builds semantic code indexes, maps dependency graphs, and performs intelligent code search across large codebases.
agentsope/SkillAlchemy
SOP for terminal-based, git-native AI pair programming with Aider (git work-tree + tree-sitter repo-map + edit-format + human-in-loop REPL).
agentsope/SkillAlchemy
Split a multi-call LM workflow by cognitive load, not by accuracy: let one strong model make the few reasoning decisions and a cheap model do the many mechanical executions (Aider architect+editor…
agentsope/SkillAlchemy
SOP for building multi-agent systems with CrewAI — role-based collaboration, sequential/hierarchical processes, Flows, memory, delegation.
agentsope/SkillAlchemy
SOP for building LLM applications on Dify — visual workflow + chatflow + agent + RAG knowledge base + plugin marketplace + observability, self-hostable.
agentsope/SkillAlchemy
Designs multiscale chunking for RAG by embedding small units for retrieval precision and returning larger context for synthesis.
agentsope/SkillAlchemy
Enhancement-overlay skill — the DECISION + WIRING layer for LM observability that the single-backend skills [[langsmith]], [[phoenix]], [[mlflow]] do NOT cover.
Categories
Coder-agent working-file budget discipline: keep the editable working set (files you /add into writable context) under ~25k tokens, separate "read" from "edit", delegate breadth to a read-only…. Agentsop Context Scope Discipline is an agent skill from agentsope/SkillAlchemy. Coder-agent working-file budget discipline: keep the editable working set (files you /add into writable context) under ~25k tokens, separate "read" from "edit", delegate breadth to a read-only repo-map, and drop files once edited.
Agentsop Context Scope Discipline fits situations like: an LLM coder-agent edits multiple files; the working set must stay focused; the model starts editing the wrong file / missing targets because too much context dilutes attention.
Run `npx skills add agentsope/SkillAlchemy --skill agentsop-context-scope-discipline -a claude-code`. Or copy the skill folder (skills/agentsop-context-scope-discipline in agentsope/SkillAlchemy) into .claude/skills/agentsop-context-scope-discipline in your project. Claude Code loads it when a task matches its description.
Run `npx skills add agentsope/SkillAlchemy --skill agentsop-context-scope-discipline -a codex`. Or copy the skill folder (skills/agentsop-context-scope-discipline in agentsope/SkillAlchemy) into .agents/skills/agentsop-context-scope-discipline 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 agentsope/SkillAlchemy --skill agentsop-context-scope-discipline -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/agentsop-context-scope-discipline, .gemini/skills/agentsop-context-scope-discipline, .github/skills/agentsop-context-scope-discipline and .opencode/skills/agentsop-context-scope-discipline in your project.
SKILL.md names no scripts, command-line tools or credentials: Agentsop Context Scope Discipline is instructions for the agent only.
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. 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.
Agentsop Context Scope Discipline is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
About 3k tokens (SKILL.md is roughly 12k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 1.3k tokens, read only when the agent opens those files.
Skills that share tags, products or a category with Agentsop Context Scope Discipline: ccc Semantic Code Search (cocoindex-io/cocoindex-code, 2.8k stars), Repomix Codebase Packer (yamadashy/repomix, 29k stars), Codebase Explorer (yologdev/yoyo-evolve, 1.9k stars) and Caveman Repository Explorer (JuliusBrussee/caveman, 111k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
agentsope (a GitHub user) maintains it in agentsope/SkillAlchemy, which has 466 GitHub stars. The repository holds 46 skills in this directory. The repository was last updated on October 9, 2026.
Source: agentsope/SkillAlchemy on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.