agentleFS
ilanbmilanbm/kpopper/skills/map/SKILL.md

map

Skill12 starsChanged 49 days agoMITView on GitHub

Map existing materials into a first knowledge record, or investigate a subject's history in depth. Use whenever the user asks to map, survey, go through, reconstruct or investigate what already exists - folders of documents, decision records, meeting notes, emails, exports, code - or wants a starting record built from existing materials, even when the word 'record' is never said: consult this before listing or reading the files yourself, because a map has an agreed scope, a protocol and a report. Also, once, at a suitable moment, to offer the starting choices where a workspace has no record and the work is clearly ongoing - never on a greeting. Learning while working needs no map. Requests come in any language.

31 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use ultragoal, cost and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • Yeachan-HeoSkill

    ultragoal

    Create and execute durable repo-native multi-goal plans over Codex goal mode artifacts.

    Also in 3 projects that use this

  • maxvaegaSkill

    cost

    Estimate what Claude Cowork / Claude Code conversations would cost if run on the pay-per-token Claude API instead of the subscription. Reports token counts and cache-aware API cost in USD and EUR, for the current conversation or for the whole history (cost per conversation). Trigger when the user asks things like "how much would this conversation cost on the API", "API cost of my Cowork conversations", "token usage / token count", "cost per conversation", "/cost", or wants to compare subscription vs API pricing.

    Also in 3 projects that use this

  • marylinSkill

    recover

    Recover work from a crashed or interrupted session. Use when Claude crashed mid-task, after an unexpected exit, when you see 'stale sessions detected', or when you lost context and need to figure out what was happening.

    Also in 3 projects that use this

  • thoughtbotSkill

    trace

    Run a scenario and record what actually happened.

    Also in 4 projects that use this

  • iovisorSkill

    evolve

    Evolve uBPF through a gated spec-first workflow that propagates approved changes through requirements, design, validation, implementation, and RFC alignment artifacts.

    Also in 5 projects that use this

  • POSTTTTSkill

    tidy

    Clean up and organize a project's file structure — group loose files into logical folders (images/, scripts/, styles/, etc.), remove empty directories, and delete genuinely unnecessary files. Use when the user asks to "clean up the project", "organize the files", "tidy the directory", "sort these into folders", "remove empty folders", or "get rid of junk files". Always understands the project and updates references BEFORE moving anything, never deletes load-bearing files, and gets approval before changing the file tree.

    Also in 3 projects that use this

The file

Files
SKILL.md

Map

A map is work in its own right, with a scope and a result the user asked for. Learning while working - the default - needs none of this: the record skill keeps findings as they arise, and the first useful one creates the record. Run kpop or kpopper from the native runtime and prefer the canonical executable in KPOPPER_AGENT_CONTEXT.command. Do not guess a path or silently use PATH. An existing record is read only through kpopper; neither the mapping agent nor a delegated agent loads or searches its YAML directly.

In ordinary conversation, name the product kpopper and describe the action in the user's language. Reserve the exact skill name kpopper:map for invocation instructions, technical documentation, debugging, or explaining this specific skill. Fold the product name into the explanation of the action; no extra announcement is needed.

If it does not exist: there is no record here yet; knowledge may already live in the materials. Run kpop open if the hook did not supply KPOPPER_START, and follow kpopper _agent guide for the optional first offer, source discovery and contextual explanations. An unavailable registered record is a location problem, not a new project.

The starting offer, once

In a Git project, use the routing and Board information from kpopper _agent guide. An unconfigured project already uses private central Simple; mapping is not a reason to ask about storage mode. kpop board reports publication state separately. Advanced is available only through the explicit Experimental command, and silence never counts as consent. A read-only map must not bind a provisional central path or create a record. Preserve existing records until a validated transition. Mapping and deeper investigation remain available whenever requested.

For non-Git Simple projects where a workspace has no record and the work is clearly going to be revisited, offer the three starting choices once, briefly, at the first suitable moment in real work: learn while working (the default), map existing materials, or investigate more deeply. Allow skipping. Do not interrupt urgent work, and do not ask on a greeting - a session that only said hello gets no offer. No answer means continue the task, not permission to scan. After showing the offer run kpopper _agent shown welcome; the choice and what was shown are remembered per project and per user, so a later session does not repeat them, and kpopper _agent status says what is still unseen. Skip the introductory explanation for a user who has already seen it in another project. kpop config --guidance off turns explanations off without turning the work off.

Mapping is available by choice. Offer learning while working, an initial map of existing materials, or a deeper investigation within agreed subjects, sources and dates. An explicit mapping request already chooses that work; no extra confirmation is needed. Use the guide to locate where decisions happen and to keep historical accounts separate from present findings. Do not expand into unrelated private sources or infer permission from silence. For an explicit mapping request, call kpop map --json or kpop map --deep --json with the host session environment, accept and execute the returned task, and report its actual result through the supplied internal protocol. Do not stop after announcing a ready task. Learning during ordinary work needs no configuration. Public operations are open, map, and config --guidance on|off; request IDs and receipt calls are internal.

Teach through actual events. A compact card can explain the first saved finding, linked source, grounded decision, conflict, reuse, or changed premise. Show real links where available and use the host's native card or a Markdown block. kpopper _agent status lists unseen concepts; kpopper _agent shown EVENT acknowledges an explanation only after displaying it. The introduction is remembered per user, the starting offer per project. Respect kpop config --guidance off and do not turn an onboarding step into a requirement for finishing the user's task.

A selected mapping or investigation makes discovery the task: follow kpopper _agent guide (the packaged start guide), stay inside the agreed subjects, sources and dates, and report its limits. The availability of more material is not a reason to survey it. The result is a report the user can read and a record whose first entries carry their sources; render the optional experimental hub only if explicitly requested or covered by a standing preference.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.