Define intended behavior and acceptance contracts. Use when building features, changing UX/API/domain behavior, requirements are fuzzy, a request needs grilling into a contract,…
Run bounded spike or sketch investigations. Use when prototyping a data model, state machine, API shape, integration, performance idea, or several UI/product variants before…
Drive behavior changes with failing checks first. Use when implementing logic, fixing bugs, changing behavior, adding edge cases, writing tests, or needing red/green proof before…
Manage Loom team worker lifecycle from spawn to retire, including ticket assignment, progress tracking, merge operations, and workspace cleanup.
Act as Team Manager for Loom team, handling ticket workflow when team commands may be unavailable
Webhook implementation and consumption patterns. Use when building webhook endpoints, receivers, or senders — covering HMAC signature verification, retry with exponential backoff,…
AI-native workflow analyzer for Loom recordings. Breaks down recorded business processes into structured, automatable workflows.
Orient and route Loom work. Use when starting or resuming a session, the request is fuzzy, repository scope or owner chain is unclear, or you need to decide which skill handles a…
Loomio integration. Manage Organizations. Use when the user wants to interact with Loomio data.
Automate Loomio tasks via Rube MCP (Composio). Always search tools first for current schemas. — from general/automation
Loon 配置、规则、策略组、节点、DNS、复写、脚本、插件、Scheme、订阅规则、Script API 等内容的说明、生成与修改技能。当用户明确提到 Loon，或在 Loon 上下文中提到配置文件转换、规则、策略组、节点筛选、DNS 映射、Rewrite、Plugin、Script、Script API、Scheme、订阅规则时，必须触发本技能。
Use when the user wants to automate, schedule, or run work autonomously/remotely — setting up a /goal, /loop, routine, channel, or long unattended run, or asking "how do I keep…
Activate loop-auditor to periodically review and improve the Self-Evolution Loop itself. Analyze alignment between skills, identify bottlenecks, weak feedback loops, and…
Routes deliverables back to the workspace — save, create actions, update intelligence — offer always, force never
Autonomous unified B″ Worker for the backlog.md board. Drives BOTH lanes from one Monitor session: basic-ready events execute kind:basic tasks in isolated git worktrees that merge…
Check token budget and run-log spend before and after a loop run. Enforces early exit when over budget or when there is no actionable work.
Classify recurring agent, tool, permission, environment, API, CI, and workflow failures into verified operational knowledge or root-cause repair incidents.
Write an exit-condition contract before letting any AI run unattended — a background agent, a scheduled loop, a "keep going until it works" task.
Use when 진안 wants to start (register) an autonomous improvement loop on the Muse repo — "루프 돌려줘", "loop 등록", "X를 계속 강화하는 루프", or just a theme to iterate on.
Design a goal-oriented agent loop, and review it for the ways loops go wrong — spinning and burning tokens, Goodhart-gaming the verifier, or running a wrong answer to completion.
Use when the user wants to identify and model growth loops in their business — detect viral, content, data, paid, and ecosystem loops, model effectiveness, and propose new loops.
Single front door for a repo's automation loops, so one routine per repo can handle every loop event instead of one routine per event.
Design the evaluation harness for an agent loop — the proof layer that makes a loop trustworthy instead of merely confident.
Use when executing an existing HOTL workflow file — reads steps, loops until success criteria met, auto-approves low-risk gates, pauses at high-risk gates.
Evaluate-Loop Step 3: EXECUTE. Use this agent to implement tasks from a verified plan. Works through plan.md tasks sequentially, writes code, updates plan.md after every task, and…
Run a spec-driven agent loop where coding tasks live as markdown specs that move through inbox → active → archive, get implemented by Claude Code or Codex, and pass a review gate…
Evaluate-Loop Step 5: FIX. Use this agent when an evaluation (plan or execution) returns FAIL. Takes the evaluator's fix list, creates specific fix tasks in plan.md, executes the…
Use quando o usuário quer um prompt PRONTO de `/loop` pra rodar uma ação longa e repetitiva até atingir um resultado mensurável, e depois parar sozinho.
Use after an iteration, bug, review, incident, or repeated failure to decide what durable artifact should change: tests, skills, docs, decision logs, issues, or nothing.
Author and optionally direct-launch one durable loop-graph run as executor, ledger, directives, supervisor, and owner handoff artifacts under a dated `.octopus` directory.
Find, compare, adapt, and design bounded AI-agent feedback loops with explicit checks, stop rules, guardrails, and handoffs.
Grill me about specs for the workflows I want to build, within this workspace.
Monitor PR checks and fix failures until green. Uses gh pr checks as the source of truth for PR-attached checks. — from carterdea/dots
Operate autonomous agent loops, monitor progress, and intervene safely when loops stall. — from gist.githubusercontent.com/deepfriedrabbit/c087955
Decides hand-vs-compiler for loop transforms (unrolling, SIMD, fusion, hoisting). Use when reviewing/authoring a hot loop or tempted to hand-optimize one.
Main loop orchestration — task queue, subagent lifecycle, state machine, error recovery. 主循環協調：任務隊列、子代理生命週期、狀態機、錯誤恢復.
specialized protocol for Loop Engineering — auditing a workspace for repeatable tasks, ranking them with the 4-Condition Test, and building self-verifying Loop Orchestration…
Evaluate-Loop Step 2: EVALUATE PLAN. Use this agent to verify an execution plan before any code is written.
Evaluate-Loop Step 1: PLAN. Use this agent when starting a new track or feature to create a detailed execution plan.
Resume cron jobs that were previously paused with /loop-pause. Reads ~/.claude/.paused-loops.json. Optionally takes a new interval argument (e.g.
Loop Returns integration. Manage Returns, Merchants, Customers, Integrations. Use when the user wants to interact with Loop Returns data.
Use when a task should repeat on a schedule or until a condition is met. Explains Claude Code's /loop taxonomy — interval loops (fixed cadence) vs.
Execute the per-wave implementation loop: implement against spec, run verification, fix failures, repeat until green — then stop and request human checkpoint.
Framework for designing, verifying, and executing autonomous loop specifications for SAP tasks (arXiv:2607.00038)
\"Show current Ralph Wiggum loop status and task progress. 顯示當前Ralph Wiggum循環狀態及任務進度。 Use when: check loop status, view task queue, loop progress, is loop running, current task\
Triage recent changes, CI failures, issues, and conversations. Produces a concise, actionable findings report suitable for a loop to consume.
驅動一個多模型分工的自動迴圈：由 Fable 擔任 orchestrator 規劃並審定升級方案，Sonnet 擔任 implementer 實際執行修改，另一支 Sonnet 擔任唯讀 observer 記錄每個動作供事後查核，唯讀 inspector-ops 在每個完成階段查核驗收；有錯就從最前端的 prompt（先 brief、再…
loop 产出变更的独立校验子代理。默认 REJECT，除非证据充分。跑测试、确认 diff 范围。在 minimal-fix 或任何实现子代理之后使用——绝不与实现者同一角色。
Run Ruflo background workers using Claude Code native /loop scheduling
Use when the user invokes `/loopd ` to start a loopd session. The argument is either `pm` (run the PM merge-orchestration round) or a worker name (run that worker's round in…
Use when installing, bootstrapping, configuring, starting, verifying, operating, or troubleshooting Looper, looperd, the looper CLI, ~/.looper config, or runtime paths; when…
Use when an approved goal, design, or implementation plan needs autonomous hardening after a first pass, especially when the user says loopfix, loop-fix, review-fix loop, keep…
Loopify integration. Manage Organizations, Users. Use when the user wants to interact with Loopify data.
Use when a loop is stuck, repeating failures, flaking, patching the wrong target, hitting an authority conflict, or burning turns without convergence.
Use when creating, maintaining, auditing, or summarizing loop trace receipts that record checks, failures, patches, rationale, evidence, and stop reasons.
LoopMessage integration. Manage Users, Organizations. Use when the user wants to interact with LoopMessage data.
Interview the user about what a loop harness should accomplish, then generate KEY_OBJECTIVE, verify_objective(), prompt functions, and the verify() gate directly into loop.sh.
Top-level workflow loops that chain multiple personas autonomously. Use *fix, *review, *impl, *qa, or *plan sprint instead of manually invoking each persona in sequence.
Loops.so integration. Manage Persons, Organizations, Deals, Activities, Notes, Files and more. Use when the user wants to interact with Loops.so data.
Generate images and videos from React + Tailwind CSS templates using the loopwind CLI.