paleo-trim-context
starsforks
How to install paleo-trim-context?
mocasus/paleo/paleo-trim-context
$npx -y skills add mocasus/paleo --skill paleo-trim-contextUse when context window is large / token cost high / long session. Proactively trim, summarize, or drop stale content to save context tokens without losing the task state.
SKILL.md
| 1 | # paleo-trim-context |
| 2 | Keep context small. Drop stale, keep task state. |
| 3 | |
| 4 | ## Rules |
| 5 | - Summarize old tool output into 1-2 lines; keep only the result that matters. |
| 6 | - Drop: resolved errors, obsolete file reads, duplicate logs. |
| 7 | - Keep: current plan, open TODO, key IDs/paths, last decision. |
| 8 | - Use session_search / notes for history instead of re-pasting. |
| 9 | - Prefer code/script over long prose explanations in context. |
| 10 | - Pre-thinking compression (biggest savings): BEFORE the model reasons, compress retrieved docs / RAG context / long file dumps into key facts + a source pointer. Drop full-doc dumps. |
| 11 | - Effort pinning: trivial task -> signal low effort, no over-think. Reuse a prior conclusion instead of re-reasoning the same thing. |
| 12 | |
| 13 | ## Scope boundary (don't collide with siblings) |
| 14 | - Context hygiene + effort pinning + pre-thinking compression = THIS skill (runs automatically during the session). |
| 15 | - `paleo-converse` = explicit conversation-turn compression + dedup (on-demand, when the user points at history or context is heavy). |
| 16 | - `paleo-summary` = one bulky artifact (log / diff / dump / stdout) -> tight intisari. |
| 17 | - Hand off to those when the target is a specific turn-block or a single artifact; don't re-implement them here. |
| 18 | |
| 19 | ## When |
| 20 | - Long session (>20 turns), big file dumps, repeated similar output. |
| 21 | - Before calling an expensive model — shrink first. |
| 22 | |
| 23 | ## Gotchas |
| 24 | - Don't trim the user's latest instruction. |
| 25 | - Keep ground truth (file paths, IDs) — summarize derivations only. |
| 26 | - If unsure what's stale, ask once. Don't guess-drop critical state. |
People who liked this also viewed
6 picksType
- lark-drive飞书云空间（云盘/云存储）：管理 Drive 文件和文件夹，包含上传/下载、创建文件夹、复制/移动/删除、查看元数据、评论/权限/订阅、标题、版本、飞书文档密级标签（secure labels）和本地文件导入。用户需要整理云盘目录、处理云空间资源 URL/token、判断链接类型/真实 token/标题，或导入 Word/Markdown/Excel/CSV/PPTX/.base 为 docx/sSkillsJul 2026
- lark-sharedUse for lark-cli setup/auth tasks: auth login/status/logout, user vs bot identity, business-domain permissions (--domain, including all/docs/drive), missing scopes, revoking authorization, or handlingSkillsJul 2026
- lark-whiteboard飞书画板：查询和编辑飞书云文档中的画板。支持导出画板为预览图片、导出原始节点结构、使用多种格式更新画板内容。 当用户需要查看画板内容、导出画板图片、编辑画板时使用此 skill。不负责：飞书云文档内容编辑（lark-doc）、文档内嵌电子表格/Base（lark-sheets / lark-base）。SkillsJul 2026
- lark-task飞书任务：管理任务、清单和任务智能体。创建待办任务、查看和更新任务状态、拆分子任务、组织任务清单、分配协作成员、上传任务附件、注册或注销任务智能体、更新任务智能体的主页数据、写入智能体任务记录。当用户需要创建待办事项、查看任务列表、跟踪任务进度、管理项目清单或给他人分配任务、为任务上传附件文件、注册注销任务智能体、更新智能体主页数据、写入任务记录时使用。SkillsJul 2026
- lark-attendance飞书考勤打卡：查询自己的考勤打卡记录SkillsJul 2026
- lark-apps妙搭（Spark/Miaoda）应用开发与托管：应用创建、本地全栈开发、云端生成迭代、创意设计（UI mockup / 可交互原型 / 线框图 / 落地页 / 仪表盘 / 幻灯片 deck / 视觉探索）、AI相关能力和飞书平台能力或者其他外部能力集成、日志/Trace/监控指标/PV/UV 查询、环境变量管理、应用角色与成员管理、自动化触发器（定时/记录变更/Webhook/飞书审批）。当用户要SkillsJul 2026
