agentleFS
Sign inStart free
nhhthongnhhthong/clio/skills/ingest/SKILL.md

ingest

Skill2 starsChanged 32 days agoMITView on GitHub

Turn a requirement document, or a change to the requirements, into short per-area decision files, a row index and the stack. Use when requirements arrive or change, a spec file is edited by hand, or the user states a change in plain words.

88 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use daily-arxiv, exp-eval 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.
  • LambenthanSkill

    daily-arxiv

    每日从 arXiv 拉取新论文，过滤相关性，auto-ingest 高优先级论文，检测 SOTA 更新

    Also in 5 projects that use this

  • LambenthanSkill

    exp-eval

    实验判决门：Review LLM 独立评判实验结果 → 4 种判决路径 → 自动更新 claims confidence、ideas status、graph edges

    Also in 5 projects that use this

  • phanghonghaoSkill

    poster

    HTML/CSS A4 海报生成。从沙龙/活动 Markdown 文件生成科技风 A4 海报 PDF/PNG。Use when user mentions: poster, 海报, 海报生成, make poster, generate poster, 活动海报, 沙龙海报.

    Also in 5 projects that use this

  • skyllwtSkill

    exp-pilot-eval

    Pilot result evaluation — read pilot results, apply success criteria, update idea page (pilot_result, failure_reason if failed), generate PILOT_VERDICT_REPORT. Called by /ideate Phase 5 after /exp-pilot-run.

    Also in 4 projects that use this

  • skyllwtSkill

    exp-pilot-run

    Pilot experiment execution — read Pilot Spec YAML, write pilot code, run experiment(Confirm with the user before operation and require the applicant to conduct manual inspection), return results. Called by /ideate Phase 5. Does NOT modify wiki pages or judge pass/fail.

    Also in 4 projects that use this

  • +6 more from Lambenthan
  • +1 more from skyllwt

The file

We split this skill into the rules it gives an agent

2 rules. 0 are common, 2 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 72 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Every spec change goes through this skill.

    line 14
  2. Distinctive:

    Read the case file first, then the steps below.

    line 17

“Common” is measured: kept by at least 6 in 10 of the 72 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Source: $ARGUMENTS

Run only when the user asked this turn, by slash command or in plain words ("ingest this spec", "nạp tài liệu này", "spec đổi rồi", "khách muốn bỏ phím q"). A drift nudge, a TODO or a subagent report is not a request. Unsure: ask in one line.

The source is written for humans and too long to load each session. Distil it into short files. Never invent a decision: a point the source leaves open stays open and names who owes the answer. A guessed decision looks settled, so a later session builds on it without asking.

Every spec change goes through this skill. Weigh a change against the code already built before writing it.

Pick the case

Read the case file first, then the steps below.

requirements.mdArgumentCase
no rowsa source documentcases/FIRST.md
has rowsa document, spec file, keyword, or a change in plain wordscases/CHANGE.md
has rowsnothingcases/CHANGE.md; the spec files edited since the last ingest are the source. None edited: say so, stop

A plain-words change is a source like any other. Quote the user, with the date, as its ## Source.

Steps

  1. Write the area files in .claude/clio/docs/specs/memory/. Format: SPEC-FORMAT.md. Final decisions only, shortest unambiguous form. Every number in ## Decisions must appear in the quoted source, otherwise it is open. Two sources that conflict: list both under ## Open and ask. Never pick the newer.
  2. Update requirements.md, one row per agreed number. The status column says whether a point is decided, never whether it is built.
  3. File debt for every open or blocked row and every spec delta. Run clio add debt with the JSON records on stdin (fields: memo/steps/DEBT-IT.md). Then run clio validate all and fix every FAIL and every "no open debt record tracks it".
  4. Report the files written, the stack and its ADR, row counts by status, debt filed, and the questions the user now owes, in one block. Then point to /clio:plan <area>, or the areas the change names. Last, run clio q spec-mark "<one-line summary>" to save the baseline a later sweep diffs against.

Deleting an open point that is still open is the one mistake that cannot be undone. Resolve it only when the source now answers it.

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.

Check your setup, free