en un clic
plugins-haiku
plugins-haiku contient 26 skills collectées depuis mwguerra, avec une couverture métier par dépôt et des pages de détail sur le site.
Skills dans ce dépôt
Use when asked to implement, build, add, fix, refactor, or change anything in a codebase — vague asks included. Tier-driven phases: intake, investigate, plan, build, verify, deliver. Entry point: /implement.
The map-reduce + blackboard pipeline that lets a small-context model digest inputs far beyond one attention window — a codebase, a long document, a corpus, a multi-part idea — and state only evidence-anchored, citation-checked conclusions. Use when / Trigger: executing or resuming an analyst run (/analyst:digest, /analyst:resume, /analyst:show), or any analysis input exceeds one 200-line chunk.
Use whenever a change needs a system-design decision — architecture, system design, data model, module/service boundaries, interface/API contracts, scale, or an ADR. The canonical discipline for designing systems: tradeoffs made explicit, the design sized to the 15 elicited context fields, one-way doors found by a five-flag checklist (one ADR each), every component earning its place through a two-cell subtraction walk, and every "start simple" seam carrying a concrete extraction trigger. Executed as checkpointed run steps (references/phase-*.md) that write into docs/architecture/ + docs/adr/ per scribe's documentation contract.
Stay focused on the project with the task in hand; recall another project's knowledge ONLY when the task actually needs it. When a task requires changing a shared dependency you own (a local composer/npm package in another repo), make the change IN that owning repo — commit, push, retag — then update and re-verify the consumer. Trigger: an edit is about to land inside a dependency directory, or a task can't be done without touching a package you maintain elsewhere.
Use when asked for a deep audit, production-readiness review, architecture review, or prioritized recommendations across a codebase or subsystem — orchestrates parallel reader agents with adversarial verification and publishes findings to docs/deep-analysis/
Run a large, multi-place task as a checkpointed state machine over the runs store — plan, adversarial plan review, wave-by-wave execution, synthesis — one step per invocation, resumable after any interruption. Trigger: cross-file refactor, data migration, large feature build, multi-source research — work that cannot finish in one pass. Manual invocation: /deep-work.
Use before ending any turn that used tools or produced a deliverable — when tempted to ask "Want me to…?", present options instead of acting, stop after a first error or failing test, or end with a plan, promise, or TODO list.
Use when writing any user-facing reply — answers, status updates, summaries, or final reports — especially after multi-step work, when tempted to show process, add headers to a short answer, open by classifying the question, or open with praise.
Use before claiming anything works, is fixed, is done, or passes; before status updates on changes you haven't run; before agreeing with someone else's claim about your work; and before state-changing commands like restarts, deletes, or config edits.
Prove a change didn't break existing work: run the suite on current source, run the tier-prescribed blast-radius probes against docs/architecture/ contracts, report coverage honestly, and emit the verdict taskmanager's done-gate enforces. Trigger: a change is about to be marked done or committed.
Turn the work just done into VETTED lessons that survive across sessions and projects — harvest the session's decisions/errors, draft candidate lessons, refute them adversarially, and promote only the survivors into taskmanager memories. The self-improvement loop, gated so a false lesson can never silently change how the suite works. Trigger: end of a substantial piece of work / a wrap-up moment, or when asked to capture lessons learned.
Use when implementing any change in existing code — when tempted to fix nearby code, add unrequested validation or options, fix something "arguably in scope", or when the diff is growing past what was asked.
RETIRED — the thoughts/ scratch-directory convention is superseded by the runs store. Trigger: any impulse to create thoughts/<task-slug>/ or to persist a multi-session board outside the database.
This skill should be used when conducting PRD interviews, creating product requirements documents, planning new features, documenting bug fixes, or when using commands like "/prd-builder:prd", "/prd-builder:feature", "/prd-builder:bugfix", or "/prd-builder:refine". Provides comprehensive interview frameworks and question templates for building thorough PRDs.
Use whenever you make or discover a decision, fix a non-trivial error, change architecture or what the system does, or defer something undecided — and before acting on any task. The canonical contract for how docs/ works: docs/ is the single source of truth (read it first, write it after), its truth is typed and verified, and stale docs are always a defect. Adopted by every plugin, command, and agent that reads or writes project state.
Manage project memories - constraints, decisions, conventions with conflict detection and resolution
Manage tasks, state, and memories - parse PRDs into hierarchical tasks with dependencies and complexity
The staged external-thinking protocol: micro-invocations that write a durable trace so a first draft cannot be delivered as the final answer. Use when / Trigger: executing or resuming a thinker run (/thinker:think, /thinker:resume, the think_trigger hook card), or any hard question needs a reasoned, auditable answer — a diagnosis, design choice, estimation, tradeoff, causal-why, or plan.
The house rules for building Filament v5 panels — schema layout, navigation grouping, relation managers, save-redirects, delete placement, the user-menu profile/settings page, native authentication + optional authenticator-app (TOTP) 2FA, and no public registration. Opinionated, prescriptive, and grounded in real Filament 5.6 source so every namespace and method is correct. Trigger: load whenever editing Laravel/Filament code — creating or changing a Filament Resource, form/table schema, page (Create/Edit/List), RelationManager, or PanelProvider; wiring panel auth/2FA; or scaffolding admin UI with `make:filament-*`.
Opinionated CORE Laravel 13 + Pest 4 conventions, applied by default on any Laravel project — artisan-first file creation, Eloquent/migration/queue/ scheduler patterns, fail-closed multi-tenancy, parallel-safe tests, and a "verified before done" bar. Composes with (never duplicates) maestro's stack-agnostic discipline. Use when / Trigger: editing or creating anything in a Laravel codebase — models, migrations, factories, jobs, scheduled commands, Filament resources, or Pest tests; whenever you reach for a file generator, design a schema/relationship, write a queued job or a tenant-scoped query, or are about to call a Laravel task "done."
The ensemble verification harness: split a conclusion, fix, or design into checkable claims, classify each to a pre-written probe row, spawn N independent voters (N from the tier: 1/3/5), and aggregate mechanically. Trigger: a conclusion/fix/design is about to ship, a core path changed, or a pipeline step names an adversarial pass. Manual invocation: /adversarial-verify.
Use at the start of any multi-step task and during exploration — before reading files, searching, or re-checking completed work, especially when tempted to read whole files, re-verify known facts, or run independent lookups one at a time.
Use when writing or editing code in an existing codebase — before adding comments, docstrings, try/catch blocks, validation, logging, or TODOs the surrounding file doesn't have, and before explaining your style choices in the reply.
The suite's front door: classify a request into one of five lanes AND read the tier from the sizing artifact — emit pipeline + tier together. Depth is a routing output, not a brain judgment (ADR-0002). Trigger: a "which pipeline / where do I start / what process for this" question, or an ambiguous ask whose right path is unclear.
Find every judgment call in a draft prompt and compile it into a form a Haiku-class model can execute: a decision table over countable inputs, an ensemble vote, an evidence artifact, or a mechanical escalation. Use when / Trigger: writing or reviewing ANY prompt text for this marketplace — especially before shipping a SKILL/command/agent containing words like assess, decide, right-size, when it matters, if needed, appropriately, thoroughly, sensibly.
How to structure a Claude Code plugin so a Haiku-class model executes it reliably — router stubs, phase files, hard prompt budgets, externalized state, verdict contracts, hook patterns, and the done-bar. Use when / Trigger: creating a new plugin for this marketplace, writing or restructuring any SKILL.md / command / agent / hook, or porting a frontier-era plugin to small-model form.