Menu
File a new work item as a card in wiki/backlog/ (v2 — card-is-spec, slug filename, proportional detail). Use whenever the user reports a bug, asks for a feature, or proposes new work mid-task — instead of inline-patching.
Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.
Basé sur la classification professionnelle SOC
Primary entry point for a freshly cloned Vibin seed. Interviews the user, seeds the knowledge graph, scaffolds the stack, writes the permission profile, strips seed-meta, and hands off to the loop skill. Use when starting a new project, or when the user says "bootstrap" or "let's begin".
Run the Vibin v2 build loop — pull a work item into the sprint, then retrieve → do → verify → capture → review → close, mostly inline. Use to build the backlog or resume work.
Bring an existing Vibin project up to date with the latest seed by diffing the project's recorded seed commit against the latest Vibin on GitHub and applying the changes. Use when the user says "migrate", "migrate-vibin", "upgrade Vibin", "apply the latest seed changes", or after a new seed release. Reads the seed commit hash from .vibin-version, audits the project's actual state, applies the migrations added since, and reconciles seed-owned files while preserving local customizations.
Run multiple sprints in parallel — partition backlog cards, create git worktrees per sprint, spawn background subagents, merge when done. Use when you want to parallelise independent concerns across concurrent sprints.
Execute a pre-composed sprint — retrieve, do, verify, capture, commit, close. No planning, no user checkpoints. Used by the loop skill after planning, and by the sprint-runner agent for parallel sprints.
One-screen v2 status — current sprint tasks + run log tail, backlog cards, knowledge-graph size. Read-only.
| name | intake |
| description | File a new work item as a card in wiki/backlog/ (v2 — card-is-spec, slug filename, proportional detail). Use whenever the user reports a bug, asks for a feature, or proposes new work mid-task — instead of inline-patching. |
| disable-model-invocation | false |
Captures new work as one card in wiki/backlog/ so the current task continues undisturbed. Safe for
any agent: it writes one new file and touches nothing else.
Name it by slug — wiki/backlog/<slug>.md (lowercase, hyphens, ~40 chars). No numeric ids:
slug filenames can't collide across branches the way B<max+1> did. If a card with that slug
exists, it's the same topic — update that card instead (one home).
Write the card — the card IS the spec. Detail is proportional: a one-line idea is fine; a bug needs what's known of the repro, the suspected fault, and a fix direction. Don't pad.
---
title: <short imperative title>
type: feature | bug | research | chore
priority: high | medium | low
flags: [] # [review] to pause for user sign-off before work
created: <YYYY-MM-DD>
---
## What / why
<the ask, in the user's voice — plus repro/fault/fix-direction for bugs>
## Notes
<links to relevant knowledge atoms (work → knowledge only), gotchas>
Link knowledge, don't restate it — if relevant atoms exist (wiki/knowledge/), link them in
Notes. Never copy atom content into the card.
GitHub-sourced items: note the issue (#N) in the card; the eventual commit says fixes #N
and the card is deleted on close.
Report back: "Filed <slug> (type, priority) in wiki/backlog/."
Rules: one item per invocation; do not start working it; the loop pulls it into the sprint later.