fabien-claude-marketplace

Shared hooks, agents, commands, skills across projects

5 skills · 5 commands · 1 agent · hooks

Astro 5.x — components, routing, collections, i18n, SEO, Tailwind, React islands, transitions

11 skills

Chantier→spec→plan→unattended run workflow for Claude Code: cut an initiative into an ordered just-in-time backlog (/goal:tickets), settle the functional contract of any source — Jira US (via MCP), GitHub issue, spec file or inline — with a grilled, observable Definition of Done (/goal:spec), lock the executable plan into gate-blocked iterations (/goal:plan), then let the runner execute it unattended, each slice judged by a program (goal-gate.ts) that runs the declared command and is the only thing allowed to commit — watched to the end by /goal:supervise, with /goal:next as the checkpoint between two iterations. GitHub, commit and PR are all opt-in, never forced.

6 skills · 5 agents

Content & copywriting playbooks — LinkedIn posts, SEO blog articles, direct-response copy, editing pass, editorial strategy, content calendar, content repurposing, SEO briefs, technical SEO audit, schema markup.

10 skills

PHP 8.0–8.3 language conventions — code style, OOP, DDD, refactoring, SQL, Composer. Framework-agnostic. Pair with phpunit for testing and symfony for Symfony-specific FormType/Twig/PRG conventions.

10 skills

TypeScript conventions, typing, DDD events, functional programming, OOP, refactoring

9 skills

Cross-language software craftsmanship principles — refactoring, DDD, OOP, TDD, testing, code style — that complement language-specific example skills (php-*, ts-*, etc.)

8 skills

Strategic marketing — ICP definition, persuasion psychology (70+ mental models), 139 marketing ideas, competitor analysis, positioning/GTM (April Dunford), shared product-marketing context.

6 skills

Certifies any skill directory against the agentskills spec (level 2), hard platform limits from VS Code Copilot and the Codex loader (level 3), and non-blocking authoring advisories (level 4) — sourced, precise, fail-closed verdicts. Ships the skill/agent/plugin authoring conventions themselves. Run with `node plugins/skills/scripts/certify.ts <skill-dir>`.

4 skills · 1 command

Personal Symfony conventions — FormType design (data_class as single source of truth, DataTransformer placement), Twig component patterns (trans_default_domain isolation), Post/Redirect/Get for HTML controllers, and testing anything that boots a Symfony kernel (WebTestCase built-in assertions, TypeTestCase, container doubles). Use alongside php (language) and phpunit (framework-agnostic testing). Distinct from atournayre/symfony — these are personal overlays.

5 skills

Docker, Drizzle ORM, GitHub Actions, pnpm workspaces, Zod schemas

5 skills

Distribution channels — multi-platform social posts (Reddit/Twitter/X/LinkedIn/IG/FB/TikTok), Twitter threads and long Reddit posts, email subject lines, newsletter growth.

4 skills

Audit hub — personal overrides for upstream audit skills + stack-specific code patterns (security for TS/NestJS, PHP planned). Loads in parallel with netresearch/security-audit.

2 skills · 1 command

Career material for the French tech market, built once and projected per target. career-evidence keeps a versioned referential of engagements (missions, audits, training) where every client name carries its own disclosure level; cv-tailor projects it into a two-page ATS-safe CV or an ESN-style dossier de compétences; linkedin-profile audits and rewrites the profile for how LinkedIn Recruiter actually retrieves and ranks. Rules are sourced from vendor documentation (Workday, Greenhouse, Ashby, Apec, LinkedIn Engineering), never from resume-industry folklore.

3 skills

Frontend clean architecture (hexagonal), Container/Presentation patterns

3 skills

Reverse-engineer an unfamiliar or legacy codebase into a versioned knowledge base (shared brain): recon + git archaeology, architecture map, use cases aggregated by actor goal, entity model, risk register. Two depths: onboarding (dev) and audit-prep (security-surface dossier feeding the security-audit skills). Plus a text-only flow cartography: flow-map inventories one application's inbound and outbound channels with a DDD context-map reading, flow-consolidate merges several projects' maps and closes each one's unknowns with another's assertions. Run via legacy:discovery, legacy:flow-map, legacy:flow-consolidate.

3 skills

Marketing analytics — tracking setup (GA4/GTM/UTM/tracking plan), Google Analytics Data API reporting, Google Search Console reporting.

3 skills

Cherry-picked mattpocock/skills: grill-me (relentless interview before code), grill-with-docs (interview + CONTEXT.md/ADR docs), zoom-out (broader-context request)

3 skills

Transverse git & PR discipline — ref freshness (fetch first), branch/commit/PR conventions, force-push and worktree guardrails. Merges the former PR-creation skill.

1 skill · hooks

macOS / BSD platform discipline — shell/bash 3.2 vs Homebrew bash 5+, BSD vs GNU command portability, common macOS pitfalls

1 skill · hooks

NestJS architectural conventions, DDD with NestJS

2 skills

PHPUnit testing discipline for PHP projects — TDD workflow (red-green-refactor with cross-layer iterations), test writing conventions (DAMP, AAA, Prophecy, factories). Layers on craft:tdd-workflow-principles and craft:testing-principles for the cross-language baseline; PHPUnit-specific commands and Prophecy examples live here.

2 skills

Product delivery discipline — slice any functional spec into thin vertical iterations (vertical-slice: 17 splitting techniques + a method to choose and combine them per feature, technical constraint and execution mode), and ship each one without blocking (delivery: feature flags sized to the project, additive no-BC-break change, expand/contract, and the cleanup that closes the loop once it's in prod).

2 skills

Security audit patterns (OWASP, CWE, CVSS) with 80+ checkpoints — by Netresearch

1 skill · hooks

Runtime security hooks: CLAUDE.md injection scanner (SessionStart) + Bash prompt injection detector (PreToolUse)

1 command · hooks

Cherry-picked obra/superpowers skills: verification-before-completion (evidence before claims), systematic-debugging (root-cause-first)

2 skills

Vitest TDD workflow, test conventions and patterns

2 skills

jQuery code conventions — module structure with exported init function, js-* selector classes, per-block scoping, symmetric toggleClass over imperative show/hide helpers. Framework-agnostic (works with any backend / build system).

1 skill

node:test conventions: declaring tests, node:assert vocabulary, mocks and timers, running TypeScript tests with node --test

1 skill

Deep comparative audit of external Claude skill packs against this marketplace, weighted by my real Claude usage profile, plus a repo structure + documentation health pass. Produces versioned gap-analysis backlogs under audits/. Run via /self-audit:compare <target>.

1 command

Bash test suites and Claude Code hook tests: one sourced harness, hermetic cases, exit-code and channel assertions, suites that cannot pass on nothing

1 skill

Claude Code statusline — cwd, git branch, model, context progress bar, 5h rate limit

1 command