adr-wizard
claudeADR lifecycle management for Claude Code. Create, supersede, deprecate, and validate Architecture Decision Records across multiple directories using a CLAUDE.md convention for ADR location discovery.
4 skills
AI-augmented development for engineers who stay in the loop. Interviews you to build a session-spec (with requirement analysis and conflict resolution), then runs an autonomous specialist agent team with back-pressure gates to implement it — while you review, guide, and approve at every milestone.
Plugin hyperloop from marketplace hyper-plugs in samrom3/claude-hyper-plugs. Commands read from the repo's manifest — 1 month ago.
hyperteamReads a session-spec, derives a task DAG, gets user approval, writes team-state.json, seeds the native task list, creates a specialist team, and monitors until the back-pressure gate passes.
hyperwork-api-scaffoldAPI scaffolding — generate route/handler/schema/test skeletons for new endpoints, following existing project patterns.
hyperwork-pythonPython tooling for impl tasks — pytest patterns, uv toolchain, mypy, fixtures. Compose with hyperwork-tdd skill for methodology.
hyperwork-tddLanguage-agnostic TDD discipline — red/green/refactor cycle. Fallback for unknown stack. Compose with language skills (hyperwork-python, hyperwork-typescript) for tooling specifics.
hyperwork-tech-writingTechnical writing — READMEs, changelogs, ADRs, docstrings, and user-facing docs. Match repo style, compress prose, preserve structure.
hyperwork-typescriptTypeScript tooling for impl tasks — vitest/ts-jest, tsc type checking, tsconfig awareness. Compose with hyperwork-tdd skill for methodology.
session-specSingle-pass session-spec generator. Outputs plans/<branch>-session-spec.md ready for /hyperteam.
hyperteam-leadCoordinator, monitor, and consult-arbiter for hyperteam runs. Handles worker questions and blockers via SendMessage — decides to unblock directly or escalate to user. Workers never contact the user directly; lead is the sole escalation path. Handles review failures, detects GATE readiness, and retur
hyperteam-reviewerReviews completed FEAT tasks and runs the back-pressure GATE check. Emits structured pass/fail results — result, task_id, and findings only. Read-only for source code.
hyperteam-workerPrimary executor that claims FEAT and DOC tasks. Loads assigned skills via `Skill` tool at claim time. Consults lead before escalating to user.
ADR lifecycle management for Claude Code. Create, supersede, deprecate, and validate Architecture Decision Records across multiple directories using a CLAUDE.md convention for ADR location discovery.
4 skills
Interactive self-referential AI loops for iterative development. Claude works on the same task repeatedly, seeing its previous work, until completion.
3 commands · hooks
Autonomous /loop-driven task completion with learning, prediction, and progress tracking
2 skills · 2 commands · 1 agent
Interactive self-referential AI loops for iterative development, implementing the Ralph Wiggum technique. Claude works on the same task repeatedly, seeing its previous work, until completion.
3 commands · hooks
Autonomous experiment loop — optimize any file by a measurable metric. 5 slash commands (/autoresearch-agent:setup, /ar-run, /ar-loop, /ar-status, /ar-resume), 8 built-in evaluators, configurable loop intervals (10min to monthly).
6 skills · 1 agent · hooks
Persistent project planning files with lifecycle hooks that inject selected plan context. Automatic recovery uses project files only. Explicit catchup modes may read same-project local session records to emit aggregate counts or bounded nonce-framed excerpts. Optional gated mode can request continuation on capable hosts and never runs commands declared in Markdown. No network upload path. Installs across 60+ agents.
6 skills · 13 commands · hooks
PRD generation and conversion skills for the Ralph autonomous agent loop
2 skills
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.