claude

forge

Adversarial multi-agent pipeline for Claude Code. Generators produce artifacts, discriminators validate them, and adversarial feedback loops drive convergence.

15 on GitHubby HatmanStackv2.2.0MITupdated 16 days agodevelopment
/

Plugin forge from marketplace claude-forge in HatmanStack/claude-forge. Commands read from the repo's manifest — 12 days ago.

What it installs

Skills (6)

  • audit

    Run any mix of repo-eval, repo-health and doc-health audits in parallel, producing one plan directory for a single /forge:run.

  • brainstorm

    Explore the codebase and turn a feature idea into a design spec (brainstorm.md) through clarifying questions; the entry point for /forge:run.

  • doc-health

    Audit documentation drift against the code across 6 phases and write doc-audit.md for /forge:run remediation.

  • pipeline

    Run the adversarial plan → implement → review pipeline on a plan directory produced by /brainstorm or an audit skill.

  • repo-eval

    Score the codebase on 12 pillars with a 3-evaluator hiring panel and write eval.md for /forge:run remediation.

  • repo-health

    Audit technical debt across 4 vectors and write health-audit.md for /forge:run remediation.

Subagents (15)

  • doc-auditor

    Documentation drift auditor (read-only). Six-phase audit of docs vs. code covering drift, gaps, stale docs, examples, links, config, and structure.

  • doc-engineer

    Documentation remediation generator. Fixes drifted docs, removes stale docs, adds stubs, and installs prevention tooling.

  • doc-reviewer

    Documentation quality gate (discriminator). Verifies doc fixes against source code and that prevention tooling works; writes feedback to feedback.md only.

  • eval-day2

    The Team Lead evaluator (read-only). Scores Test Value, Reproducibility, Git Hygiene, and Onboarding for the repo-eval hiring panel.

  • eval-hire

    The Pragmatist evaluator (read-only). Scores Problem-Solution Fit, Architecture, Code Quality, and Creativity for the repo-eval hiring panel.

  • eval-stress

    The Oncall Engineer evaluator (read-only). Scores Pragmatism, Defensiveness, Performance, and Type Rigor for the repo-eval hiring panel.

  • final-reviewer

    Final integration reviewer (discriminator). Holistic cross-phase review producing a Production Readiness Dashboard with a GO/NO-GO verdict.

  • health-auditor

    Technical-debt auditor (read-only). Assesses architectural, structural, operational, and hygiene debt with file:line severity findings; prescribes no fixes.

  • health-fortifier

    Additive remediation generator. Adds linting, pre-commit hooks, type strictness, and CI gates to lock in a clean state.

  • health-hygienist

    Subtractive remediation generator. Removes dead code, extracts secrets to env vars, and prunes unused dependencies to shrink the codebase.

  • health-reviewer

    Repo-health quality gate (discriminator). Reviews hygienist and fortifier work via tag-selected checklists; writes feedback to feedback.md only.

  • implementer

    Implementation engineer (generator). Implements a single plan phase via TDD with atomic conventional commits.

  • plan-reviewer

    Plan quality gate (discriminator). Adversarially reviews implementation plans for deadlocks, hallucinated files, ambiguity, and verifiability; writes feedback to feedback.md only.

  • planner

    Planning Architect (generator). Creates phased implementation plan files in docs/plans/<id>/ from a brainstorm or audit intake. First stage of the Forge adversarial pipeline.

  • reviewer

    Code quality gate (discriminator). Verifies a phase implementation against its spec and Phase-0 conventions using tools; writes feedback to feedback.md only.

View source on GitHub

Similar plugins

Matt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).

39 skills

mattpocock/skills 323,108

superpowers

claudecodex

Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques

15 skills · hooks

obra/superpowers 316,836

context7

claudecodex

Up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your coding agent's context.

1 MCP server

upstash/context7 64,709

Build powerful, stateful CLI interfaces for any GUI application using the cli-anything harness methodology.

5 commands

HKUDS/CLI-Anything 58,665

5 skills + 35 modes + Material Passport pipeline. Includes v3.6.7 cross-model audit gate and v3.6.8 generator-evaluator contract.

5 skills · 16 commands · 4 agents · hooks

Imbad0202/academic-research-skills 56,698

Automates browser interactions for web testing, form filling, screenshots, and data extraction

12 skills

vercel-labs/agent-browser 50,373

Maintain forge?

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 pluginslisted