scope-task
Skill2 starsChanged 6 months agoMITView on GitHub
Pour UNE tâche décrite, liste quelles rules, skills et docs ouvrir — progressive disclosure, anti-surcharge contexte.
- Reads credentials
In this repository32 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- apexSkill
- api-genSkill
- assets-pipelineSkill
- brainstormSkill
- debugSkill
- deep-updateSkill
- deploy-checkSkill
- depsSkill
- docs-genSkill
- e2e-testsSkill
- env-setupSkill
- featureSkill
- fixSkill
- forkSkill
- meta-promptSkill
- migrateSkill
- onboardSkill
- oneshotSkill
- perfSkill
- prdSkill
- qualitySkill
- refactorSkill
- reviewSkill
- saas-initSkill
- schema-genSkill
- session-startSkill
- stripeSkill
- ui-auditSkill
- uiSkill
- variationsSkill
The file
Files
Scope Task — Charger le bon contexte
Objectif : éviter d'empiler rules, skills et fichiers inutiles.
Entrée
Une phrase décrivant la tâche (ex. « ajouter endpoint factures », « corriger bug login »).
Mapping automatique rules → domaine
| Mots-clés dans la tâche | Rules à charger |
| endpoint, route, API, handler | api-routes.md + security.md |
| schema, migration, table, colonne | db-schema.md |
| login, auth, session, token, permission | auth.md + security.md |
| composant, page, UI, form, button | ui-components.md |
| test, spec, vitest, playwright | tests.md |
| .env, config, variable, secret | config.md + security.md |
| image, SVG, audio, vidéo, asset | media-apis.md |
| deploy, prod, staging, vercel | security.md |
Mapping automatique skill → complexité
| Complexité | Critère | Skill |
| Trivial | 1 fichier, bug évident | /fix ou /oneshot |
| Simple | 2-3 fichiers, feature claire | /api-gen / /ui gen / /schema-gen |
| Moyen | 4-8 fichiers, plusieurs couches | /feature |
| Lourd | >8 fichiers ou architecture | /feature + EnterPlanMode |
Sortie (format fixe)
## Tâche
[reformulation courte — 1 ligne]
## Contexte minimal
- **Rules** : [fichiers à charger selon mapping ci-dessus]
- **Skill principal** : [1 seul skill]
- **Fichiers à lire en premier** : [chemins ou patterns]
## À ne PAS charger
[rules hors sujet, skills lourds inutiles — liste explicite]
## Prochaine commande
`/[skill] [args]` — [1 ligne de justification]Règles
- 1 skill principal sauf orchestration explicite (
/prd exec). - Préférer
file:linepointers plutôt que coller du code dans le chat. - Si doute entre
/fixet/feature→/fixd'abord (plus rapide, fail-safe). - Charger les rules avant de modifier le premier fichier, pas après.
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
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.

