claudecodex

deep-loop

Loop Engineering control plane over the deep-suite — discovers work, routes to sibling deep-* plugins as maker/checker episodes, keeps durable lock-safe loop state, and hands off to fresh sessions autonomously with proof-gated terminal states and proposal-only external actions. Version 1.28.0 pins its Codex plus Agy review recovery profile, now named codex-agy-static, to current models.

9 on GitHubby Sungmin Chov1.28.0MITupdated yesterdayCoding
/

Plugin deep-loop from marketplace claude-deep-suite in Sungmin-Cho/deep-suite. Commands read from the repo's manifest — today.

This plugin runs code on your machine — its hooks fire automatically on events like tool use. The exact commands are listed below; read them before installing.

What it installs

Skills (11)

  • deep-loop-ack

    deep-loop acknowledge — marks an episode/diff as human-reviewed, reducing comprehension debt so the loop can fan out new work. Triggered by '/deep-loop-ack', 'ack the review', 'mark reviewed', 'I reviewed it', '검토 완료', '리뷰 확인', '이해 표시', cross-platform Skill({ skill: \"deep-loop:deep-loop-ack\" }).

  • deep-loop-compact

    Prepare or restore a deep-loop workstream-session checkpoint across Claude and Codex compaction. Use for '/deep-loop-compact prepare|restore', '$deep-loop:deep-loop-compact prepare|restore', compact preparation, compact restore, context compression, 압축 준비, or 압축 복원.

  • deep-loop-continue

    deep-loop main tick — advances the kernel-returned next action in the current Workstream owner conversation, using native compact in place and handing off only at an exact terminal Workstream boundary. Triggered by '/deep-loop-continue', 'continue the loop', 'advance the loop', 'next tick', '루프 진행',

  • deep-loop-discover

    deep-loop manual discovery heartbeat — surveys the repo, git state, sibling artifacts, and existing loop state to find candidate work items, then persists them. Triggered by '/deep-loop-discover', 'discover work', 'find next work', 'what should I do next', '할 일 발견', '작업 발견', '다음 할 일 찾기', cross-platf

  • deep-loop-finish

    deep-loop finish — end-of-work: writes the final report, then transitions the run to completed (proof-gated) or stopped, and delegates to deep-memory / deep-wiki when installed. Triggered by '/deep-loop-finish', 'finish the loop', 'wrap up', 'end the run', '루프 종료', '작업 마무리', '런 종료', cross-platform S

  • deep-loop-handoff

    deep-loop boundary handoff — follows the kernel next-action and emits a fresh-owner handoff only for the exact first terminal Workstream boundary. Triggered by '/deep-loop-handoff', '$deep-loop:deep-loop-handoff', 'hand off at the boundary', 'emit handoff', '인수인계', '워크스트림 핸드오프', cross-platform Skill

  • deep-loop-resume

    deep-loop resume — validates a kernel-published boundary, affinity-recovery capsule, or project-root relocation descriptor, acquires only through its exact route, and delegates worktree entry to continue. Triggered by '/deep-loop-resume', '$deep-loop:deep-loop-resume', 'resume the loop', 'take over

  • deep-loop-status

    deep-loop status — shows the current run's status, budget, comprehension debt, circuit breaker, pending human reviews, session chain, and workstreams. Read-only. Triggered by '/deep-loop-status', 'loop status', 'show the loop', 'where are we', '루프 상태', '상태 보기', '진행 상황', cross-platform Skill({ skill:

  • deep-loop-triage

    deep-loop triage — classifies discovered candidates into actionable / needs_human / blocked / archived. Triggered by '/deep-loop-triage', 'triage work', 'classify candidates', '작업 분류', '후보 분류', '트리아지', cross-platform Skill({ skill: \"deep-loop:deep-loop-triage\" }).

  • deep-loop-workflow

    deep-loop 비공개 코어 워크플로우 — 프로토콜 adapter 4-verb(dispatch/awaitResult/checker/readArtifacts) 수행법, 리뷰 전략 조립, 자율 handoff/respawn 호출 규약을 정의한다. deep-loop 진입·continue 스킬이 references로 로드한다.

  • deep-loop

    Use when starting a durable long-running goal or coordinating multiple workstreams. Triggers: '/deep-loop \"<goal>\"', 'start a loop', 'loop engineering', 'orchestrate this work', '루프 시작', '딥루프 시작', '장기 목표 실행', cross-platform Skill({ skill: \"deep-loop:deep-loop\", args: \"<goal>\" }).

Hooks (3)

  • PreCompact

    node -e "const{join}=require('node:path');const{pathToFileURL}=require('node:url');const r=process.env.CLAUDE_PLUGIN_ROOT||process.env.PLUGIN_ROOT;if(!r){console.error('deep-loop: plugin root unavaila

  • PostCompact

    node -e "const{join}=require('node:path');const{pathToFileURL}=require('node:url');const r=process.env.CLAUDE_PLUGIN_ROOT||process.env.PLUGIN_ROOT;if(!r){console.error('deep-loop: plugin root unavaila

  • SessionStart

    node -e "const{join}=require('node:path');const{pathToFileURL}=require('node:url');const r=process.env.CLAUDE_PLUGIN_ROOT||process.env.PLUGIN_ROOT;if(!r){console.error('deep-loop: plugin root unavaila

View source on GitHub

More from Sungmin-Cho/deep-suite

All 10 plugins →

deep-work

claudecodex

Evidence-Driven Development Protocol — a single-command Brainstorm → Research → Spec → Plan → Implement → Test → Integrate workflow with TDD enforcement, governed specification and methodology authority, adaptive risk-based review and verification, per-phase model selection (Claude Code / Codex), and a hard separation between planning and coding.

27 skills · 4 agents · hooks

9

deep-memory

claudecodex

Cross-project semantic memory — harvests sibling-plugin artifacts into reusable memory cards (rule-based plus LLM distillation) and surfaces task-specific briefs, with opt-in cross-runtime hook capture and multi-pass redaction.

7 skills · 1 agent · 1 MCP server · hooks

9

deep-review

claudecodex

Independent Evaluator for AI coding agents. v2.11.2 retires Grok CLI 1.0.4, admits 1.0.13/1.0.52, and checks pinned CLI version/help through native helpers; grok-4.7 stays pinned and main stays binary-free.

4 skills · 1 command · 2 agents

9

deep-wiki

claudecodex

Blocked terminal-prune diagnostics with a preserve-first quarantine plan and journaled wiki mutation

6 skills · hooks

9

deep-evolve

claudecodex

Autonomous Experimentation Protocol — a measured outer loop that proposes, runs, and selects code changes against a fitness metric, with lineage and session history, for Claude Code and Codex. v3.6.3 parses Grok PreToolUse envelopes in the readonly guard so Grok file and shell tools are classified instead of denied as malformed.

2 skills · 2 agents · hooks

9

deep-docs

claudecodex

Document gardening + authoring — v1.6.1 fixes the native Windows atomic-write identity false-positive (path-lstat dev:0); scan/garden/audit for Claude Code and Codex.

2 skills · 2 agents

9

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

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

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

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

Maintain deep-loop?

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 plugins10 plugins