How do I install the Ki Housekeeping Chatgpt skill?

Run npx skillmds@latest add knowledgeislands/ki-housekeeping-chatgpt in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Ki Housekeeping Chatgpt safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Ki Housekeeping Chatgpt?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Ki Housekeeping Chatgpt free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Ki Housekeeping Chatgpt?

knowledgeislands (@knowledgeislands) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Ki Housekeeping Chatgpt by knowledgeislands

Name: Ki Housekeeping Chatgpt
Author: knowledgeislands

Ki Housekeeping Chatgpt

Governs safe acquisition of installed ChatGPT session material through opaque, read-only local-store evidence. Use for "acquire ChatGPT sessions", "import ChatGPT sessions", or "audit ChatGPT sessions". It guides discovery, listing, faithful source reads, checkpoints, staging, and later harvest without treating source-session deletion as part of acquisition.

knowledgeislands 2b48f57 9 files · 21.3 KB Updated Sep 17, 2026

File contents

ChatGPT session acquisition

Use the provider-neutral lifecycle: acquire → stage → harvest → durable knowledge → archive/delete source.

mcp-housekeeping-chatgpt exposes read-only chatgpt_sessions_discover, chatgpt_sessions_list, chatgpt_session_read, and chatgpt_sessions_checkpoint operations over the configured installed-app store. Its *.data records are opaque: discovery returns identity, provenance, timestamps, size, and hash; read returns exact bytes as base64 without claiming decoded conversation content.

ki space acquire chatgpt import owns repository-context staging and checkpoint persistence. The MCP never writes KI state, changes the ChatGPT store, decrypts a private format, archives, or deletes a source session.

Operating modes

Mode HELP

Explain this boundary and stop without reading or changing any source session.

Mode AUDIT

Run ki repo audit --skill ki-housekeeping-chatgpt for the configured bounded local-store evidence. Missing, unreadable, or unsafe source evidence is unavailable or failing evidence, never a fallback to arbitrary paths.

Mode CONFORM

There is no local-store conform action. Acquisition and any later source-retention decision are separate, explicitly authorised operations.

Mode EDUCATE

Explain the opaque-store boundary, the four comparable provider operations, and the separation of provider reads from KI staging and harvest.

Mode REFRESH

REFRESH writes only the canonical ki-housekeeping-chatgpt source in ki-agentic-harness; when invoked from an installed copy, stop and redirect to the Harness. Revalidate that discovery remains path-bounded, records remain opaque unless documented otherwise, and source mutation stays unavailable.

Off-ramps

Durable knowledge promotion belongs to Arcadia's acquisition lifecycle.
Repository staging belongs to ki space acquire chatgpt import in tools-ki.
Archive or delete requires a later verified acquisition and harvest decision.

---name: ki-housekeeping-chatgptdescription: Governs safe acquisition of installed ChatGPT session material through opaque, read-only local-store evidence. Use for "acquire ChatGPT sessions", "import ChatGPT sessions", or "audit ChatGPT sessions". It guides discovery, listing, faithful source reads, checkpoints, staging, and later harvest without treating source-session deletion as part of acquisition.---# ChatGPT session acquisitionUse the provider-neutral lifecycle: **acquire → stage → harvest → durable knowledge → archive/delete source**.`mcp-housekeeping-chatgpt` exposes read-only `chatgpt_sessions_discover`, `chatgpt_sessions_list`, `chatgpt_session_read`, and `chatgpt_sessions_checkpoint` operations over the configured installed-app store. Its `*.data` records are opaque: discovery returns identity, provenance, timestamps, size, and hash; `read` returns exact bytes as base64 without claiming decoded conversation content.`ki space acquire chatgpt import` owns repository-context staging and checkpoint persistence. The MCP never writes KI state, changes the ChatGPT store, decrypts a private format, archives, or deletes a source session.## Operating modes### Mode HELPExplain this boundary and stop without reading or changing any source session.### Mode AUDITRun `ki repo audit --skill ki-housekeeping-chatgpt` for the configured bounded local-store evidence. Missing, unreadable, or unsafe source evidence is unavailable or failing evidence, never a fallback to arbitrary paths.### Mode CONFORMThere is no local-store conform action. Acquisition and any later source-retention decision are separate, explicitly authorised operations.### Mode EDUCATEExplain the opaque-store boundary, the four comparable provider operations, and the separation of provider reads from KI staging and harvest.### Mode REFRESHREFRESH writes only the canonical `ki-housekeeping-chatgpt` source in `ki-agentic-harness`; when invoked from an installed copy, stop and redirect to the Harness. Revalidate that discovery remains path-bounded, records remain opaque unless documented otherwise, and source mutation stays unavailable.## Off-ramps- Durable knowledge promotion belongs to Arcadia's acquisition lifecycle.- Repository staging belongs to `ki space acquire chatgpt import` in `tools-ki`.- Archive or delete requires a later verified acquisition and harvest decision.

Frequently asked questions