claude

goal

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.

9 on GitHubby FabienSallesv3.9.3MITupdated 1 month agoproductivity
/

Plugin goal from marketplace fabien-claude-marketplace in FabienSalles/claude-marketplace. Commands read from the repo's manifest — 1 month ago.

What it installs

Skills (6)

  • grill-adversarial

    Opt-in adversarial plan-challenging pass for the goal workflow (invoked by /goal:spec when functional gaps are suspected, or on explicit request). Composes the Pocock grill skills (grill-me, grill-with-docs) and goes much further, hunting the functional and technical holes that survive a normal gril

  • next

    Checkpoint between two /goal iterations — verify the finished iteration's Definition of Done, reconcile the plan with what actually changed in the codebase, verify the working tree is safe for a fresh session (in manual mode never stage — the tree is left intact for the developer's review; staging/c

  • plan

    Executable contract of the goal workflow (Session 1) — from a functional contract built by /goal:spec, or any source whose functional level is already settled, surface the technical consequences, map every business rule to a command-line check, decide delivery mode / commit policy / remote, decompos

  • spec

    Functional contract of the goal workflow — normalize any planning source (Jira US, GitHub issue, spec file, PRD/BMAD story, a ticket cut by /goal:tickets, brainstorm, inline note), grill the functional gaps closed one question at a time, build the functional Definition of Done (an observable accepta

  • supervise

    Launches `node goal-run.ts` with a plan path in the background and watches it to the end. On a non-zero exit it classifies before repairing — plan at fault, implementation at fault, or unknown — never the reverse. A plan fault is repaired within a closed set (declared paths, max_diff, a mistyped pat

  • tickets

    Chantier level of the goal workflow — split an initiative too big for one spec into an ordered backlog of outcome-sized tickets, elaborated just-in-time — ticket 1 gets a handoff, the rest deliberately stay one-liners. Re-run on an existing backlog to reconcile it after a ticket ships and arm the ne

Subagents (5)

  • goal-run-auditor

    Audits a run just closed by goal-run.ts: duration and exit code per stage, what halted it, and what recurs across previous runs. Writes one report under <runs>/ and never changes code. Examples: <example>Context: goal-run.ts reached the end of a plan, landed or halted. assistant: 'I'll use the goal-

  • goal-run-implementer

    Implements one iteration of a locked goal plan for goal-run.ts, test-first, inside the paths its gate block declares. Cannot commit, push, stage or tick a checkbox. Examples: <example>Context: goal-run.ts reached iteration 5 of a locked plan. assistant: 'I'll use the goal-run-implementer agent to im

  • goal-run-lens

    Refutes the iteration(s) goal-run.ts just landed and returns one-sentence verdicts with file:line anchors. Read-only, advisory, never blocks a run. Examples: <example>Context: goal-run.ts landed every requested iteration and the global Definition of Done passed. assistant: 'I'll use the goal-run-len

  • goal-run-reviewer

    Reviews the pull request goal-run.ts just marked ready: design, error handling, security posture, and project conventions — the reading a gate is not built to give. Returns its review as text by default, reaching the developer through the run log only; posts to GitHub, banner-first, only when the br

  • goal-session-auditor

    Reads the transcripts a run left behind — the implementer's, the lens's, the reviewer's and the supervising session's own — and reports only findings anchored to a tool-call sequence. Never judges whether the work is correct; the gate already did. Examples: <example>Context: /goal:supervise watched

View source on GitHub

More from FabienSalles/claude-marketplace

All 32 plugins →

common

claude

Shared hooks, agents, commands, skills across projects

5 skills · 5 commands · 1 agent · hooks

9

astro

claude

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

11 skills

9

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

9

php

claude

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

9

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

9 skills

9

craft

claude

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

8 skills

9

Similar plugins

caveman

claude

Talk like caveman. Cut filler. Keep technical details explicit.

22 skills · 1 command · 3 agents · hooks

JuliusBrussee/caveman 119,462

claude-mem

claudecodex

Persistent memory for Claude Code: the next session starts where the last one left off.

29 skills · 1 MCP server · hooks

thedotmack/claude-mem 116,618

Taste Skill

claudecodex

tasteskill 2.5.1: professional company websites at the figma.com and x.ai level, with one section per screen, real photographs, real product UI, effects kept for tech and gaming pages, and a real-browser checker and finish gate.

1 skill

Leonxlnx/taste-skill 114,391

career-ops

claudecodex

Open-source AI job search that runs locally in your AI coding CLI: scan job portals, evaluate listings into a structured A-H report with a 1-5 score, tailor your CV, track applications

5 skills · hooks

career-ops-hq/career-ops 82,253

mem0

claudecodex

Cross-session memory and token savings for coding agents.

6 skills · 1 agent · 1 MCP server · hooks

mem0ai/mem0 68,893

pake

claudecodex

Package any website or local web build into a lightweight desktop app with Pake CLI.

1 skill

tw93/Pake 63,463

Maintain goal?

This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.

claude & codex plugins32 plugins