agentleFS
loursCreatifloursCreatif/ours-stack/study-status/SKILL.md

study-status

Skill1 starsChanged 4 months agoMITView on GitHub

Read-only study dashboard — where each study stands (criteria checked, artifacts produced, suggested next step). Scans studies/*/brief.md in the current repo; optionally includes ~/.ours-stack/studies-index.jsonl for cross-repo studies. Use when the user asks "study status", "où j'en suis", "statut de mes études", "tableau de bord", "what's next", "what should I do next" (status view, not strategy debate), or runs /study-status. Never creates or modifies files.

2 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions2 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 2

    1. kleyver14/study-skills★ 0same name
    2. Slava-code/slava-stack★ 0same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository12 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 1 is distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 4

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Criteria parsing: only lines matching ^- \[[ x]\] inside ## Success criteria until the next ## heading.

    line 87
  2. Very specific:

    Do not count brief.md as an artifact.

    line 114
  3. Very specific:

    If estimated deadline < today (from user_info Today's date) → append ⏰ to the study row in Prochaine étape and mention in bullets ("time …

    line 128
  4. Very specific:

    Keep wedge ≤80 chars with … if truncated.

    line 157
  5. Distinctive:

    No motivational language. No recommendations beyond the routing table.

    line 177

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “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

Study Status

Read-only dashboard — one glance at every local study: wedge, success criteria progress, artifacts on disk, and the next skill the routing table suggests. No file writes, no research, no summary theater.

Time budget: ~1–2 minutes. Scan briefs + artifact presence only — do not read notes.md or anchor sources in depth.

Hard rules

  • Read-only — never Write, Edit, or create files. Output lives in chat only.
  • No summary theater — table + 2–3 bullets. No essay, no "great progress!" filler.
  • Routing fidelity — next-step suggestions must follow AGENTS.md routing (see below). Do not auto-chain skills.
  • Wedge in one line — truncate ## Narrow wedge to ≤80 chars for the table; full wedge only if user asks.
  • AskUserQuestion — one question per call; title Study Status — {label}.

When to use (vs other skills)

SkillRole
/study-statusThis skill — status table + next step per study
/councilLive round table — debate the topic with a multi-figure panel
/source-scoutFind sources — does not report status
/bear-hoursFrame a new study — not a dashboard

Not auto-chained from any other skill.

Step 0: Scope

Default: current workspace — Glob studies/*/brief.md.

If no briefs found → say so; suggest /bear-hours for a new topic. Stop unless user names another repo path.

Optional global index — after listing local studies, AskUserQuestion:

Title: Study Status — Cross-repo?

Ask: "Inclure les études d'autres repos via ~/.ours-stack/studies-index.jsonl ?"

Options:

  • Oui — afficher slug, titre, repo, chemin (sans ouvrir les briefs externes)
  • Non — repo courant seulement
  • other

If yes → Read ~/.ours-stack/studies-index.jsonl (or $OURS_STACK_HOME/studies-index.jsonl if set). For each line, parse JSON fields: slug, title, wedge, repo, path. Skip entries whose path is already in the local scan. Do not open external briefs except top 2–3 most relevant (same wedge keyword overlap or user-named slug).

If user names one slug → scan only studies/<slug>/brief.md (still list others only if they ask "all studies").

Step 1: Discover studies

# List local study folders with briefs
Glob studies/*/brief.md

For each match, record slug from the path (studies/<slug>/brief.md).

Step 2: Extract brief fields

Read each studies/<slug>/brief.md. Extract:

FieldHow
TitleFirst # heading line
Slug**Slug:** line, or dirname
Created**Created:** line (ISO date)
Time box**Time box:** line (free text)
WedgeFirst substantive line under ## Narrow wedge (bold unit line preferred)
CriteriaAll - [ ] and - [x] under ## Success criteria — count checked/total
Sources## Source material — TBD if section missing, empty, or contains only "TBD" / "à scout" / placeholder; OK if anchor + core list present

Criteria parsing: only lines matching ^- \[[ x]\] inside ## Success criteria until the next ## heading.

Source material heuristic:

StateCondition
TBDSection absent, or only placeholders ("TBD", "à définir", "run /source-scout")
OKHas ### Read first or ### Core or ### Anchor with at least one linked source
PartialSome sources listed but no clear anchor — note in table, suggest /source-scout refresh

Step 3: Detect artifacts

For each studies/<slug>/, Glob or Bash test -f / test -d against the AGENTS.md contract:

ArtifactPathLabel
Notesnotes.mdnotes
Councilcouncil.mdcouncil
Dialoguedialogue/*/ (any subfolder with dialogue.md)dialogue
Layoutreport.htmlhtml
Mind mapmind-map.json or mind-map.htmlmind-map
Infographicvisual-proof.png, visual-proof.spec.md, or infographic-prompt.mdvisual
Memory palacememory-palace.json or memory-palace.htmlpalace
Deep research linkresearch.md or sources-index.mdresearch

Display: compact comma-separated labels in the Artefacts column (e.g. notes, council, mind-map). Use — if none beyond brief.md.

Do not count brief.md as an artifact.

Step 4: Time box signal

Parse Created as YYYY-MM-DD. Interpret Time box heuristically:

Time box textDeadline estimate
"aujourd'hui", "one session", "today"Created + 0 days
"une semaine", "1 week"Created + 7 days
"deux semaines", "2 weeks"Created + 14 days
explicit date in textuse that date
unclearskip overdue check

If estimated deadline < today (from user_info Today's date) → append ⏰ to the study row in Prochaine étape and mention in bullets ("time box dépassée").

Step 5: Suggest next step (routing)

Apply in order per study — first match wins:

ConditionSuggested step
No brief.md/bear-hours
Sources TBD or Partial/source-scout
Sources OK, no notes.md/dense-read (anchor from brief)
notes.md present, criteria not all checkedContinue /dense-read — or /council to debate the topic if stuck
All criteria checkedÉtude terminée — archiver, ou nouveau wedge via /bear-hours
User asked "what should I do next?" on one study with real ambiguityThis skill shows facts first; suggest /council only if the user wants to debate the topic

Optional artifacts (never auto-suggest unless user history implies it): /mind-map, /layout-html, /infographic, /memory-palace, /dialogue — mention in bullets only if criteria nearly done and no visual artifact yet.

Step 6: Output format

Main table

| Étude | Wedge | Critères | Artefacts | Prochaine étape |
|-------|-------|----------|-----------|-----------------|
| robotique-assemblage-structurel | comment un robot pose des éléments structurels… | 0/3 | — | `/dense-read` |
| biomimetisme-locomotion-chantier | locomotion pattes sur terrain instable… | 2/3 | notes, council, mind-map, visual, palace | `/dense-read` |
  • Étude column: slug (link path studies/<slug>/brief.md if the UI supports it, else plain slug).
  • Keep wedge ≤80 chars with … if truncated.
  • Prochaine étape: one skill or short phrase; add ⏰ prefix if time box overdue.

Cross-repo appendix (if index included)

Separate small table — no wedge/criteria unless brief was opened:

| Slug | Titre | Repo | Chemin |
|------|-------|------|--------|

Synthesis bullets (2–3 only)

Pick concrete facts, e.g.:

  • Closest to done: <slug> — N/M criteria, artifacts: …
  • Stalled: <slug> — sources TBD since Created, or time box ⏰
  • Dormant: studies with Created >14 days ago and no notes.md

No motivational language. No recommendations beyond the routing table.

Example invocation

/study-status
/study-status robotique-assemblage-structurel
/où j'en suis sur mes études

Failure modes

SituationResponse
No studies/ in workspace"Aucune étude locale. Lance /bear-hours pour cadrer un sujet."
Slug not foundList available slugs from Glob; ask once
studies-index.jsonl missingNote file absent; local scan only
Malformed brief (no wedge)Show row with wedge —; suggest brief cleanup via /bear-hours revisit

Privacy

Status output may echo wedge text and criteria from briefs. Do not paste long quotes from notes.md or copyrighted source titles beyond what the brief already lists.

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.