cc-repete
claudeSelf-evolving autonomous loops for Claude Code — a Stop-hook loop engine that drives an agent toward a verifiable goal with an evolving payload, checkpoint gating, and a context budget.
1 skill · 4 commands · hooks
Chief-operator charter for Claude Code sessions — materializes an operating charter (solo + orchestrated modes), an append-only evidence ledger, and a Stop-hook completion gate into a project via /cc-operator:start.
Plugin cc-operator from marketplace betmoar in betmoar/ccp-market. Commands read from the repo's manifest — 1 month ago.
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.
chief-operatorUse when the user asks to run as chief operator, orchestrate this, operator mode, or set up an engagement — routing to the operator charter and its start command.
holdoutUse when deriving, repairing or re-deriving a holdout — a check suite written from the spec by something that cannot read the code — or when a holdout check disagrees with the system, or when tempted to hand-edit one.
/cc-operator:brainstormRun the brainstorm workflow — N divergent directions, a blindspot scan of this codebase, and a reference search, converged into ranked options and the questions only you can answer.
/cc-operator:crawlRun the crawl workflow — one cheap crawler seat per shard in parallel, then a judgment-tier merge — to digest a large corpus without paying judgment tier for the reading.
/cc-operator:debateRun the debate workflow — 2-5 named models argue the same case over three blind rounds, then a neutral synthesis. It never picks a winner; you decide.
/cc-operator:handoffRun when ending an operated engagement or session — to produce the structured operator→human handoff before you stop or hand off.
/cc-operator:implementRun the implement workflow — one implementer seat per task, serially, on the IMPLEMENT tier, refusing an incomplete dispatch packet before spending a seat.
/cc-operator:planRun the plan workflow — decompose an approved spec into TDD tasks, then vet each in parallel for feasibility and testability.
/cc-operator:reviewRun the review workflow over an artifact — parallel narrow lenses then an adversarial verifier — with the tier bindings resolved for you.
/cc-operator:specWrite, check and approve the engagement's spec — the artifact that carries an approved design from brainstorm into plan, with a source stamp and a ledger row.
/cc-operator:startRun at the start of a session you will operate as chief operator — orchestrating an engagement, gating on evidence, or driving multi-step work to a verified done-state.
/cc-operator:tiersResolve tier→model bindings, apply one-off overrides, or render project-layer agents so plain Agent dispatch can run on a configured cc-proxy model. Wraps the resolver (ops-tiers.sh) and renderer (ops-render.sh) — the command adds no logic; those scripts' charset guard is the validation.
/cc-operator:tutorialWatch the evidence gate work — a throwaway project where the real Stop hook blocks on an open task and then lets the stop through once a verdict row records the evidence.
op-authorJudgment-tier implementer for prose, design, and any work whose quality depends on taste or reasoning, dispatched by the operator.
op-brainstormDivergent idea generator — many genuinely distinct candidate approaches, names, or angles. Generation, not judgment; a stronger model evaluates and chooses. Dispatched by the operator or the brainstorm workflow.
op-crawlerOne-shard code crawler for large-corpus fan-outs — reads the assigned shard (explicit paths) and returns a dense, faithful digest. Read-only. Does NOT fan out; the crawl workflow owns sharding and merging. Dispatched by the crawl workflow or the operator.
op-debaterRead-only debate seat — holds ONE position across rounds, engages rival positions on their strongest form, and revises only on evidence. Dispatched by the debate workflow.
op-mechanicMechanical-tier implementer for scaffolds, verbatim-specified files, fixtures, verification commands, commits, and reverts, dispatched by the operator.
op-reviewerJudgment-tier read-only reviewer and scorer, dispatched by the operator. Runs in spec mode, quality mode, or scoring mode as set per dispatch.
op-scoutRecon-tier read-only scout for searches, lookups, and "where/how is X" questions that need no judgment, dispatched by the operator.
op-verifierFresh-context adversarial verifier, dispatched by the operator after implementation work — independently tries to refute the claimed outcome and returns CONFIRMED or REFUTED with evidence.
bash "${CLAUDE_PLUGIN_ROOT}/scripts/ops-sessionstart-hook.sh"
bash "${CLAUDE_PLUGIN_ROOT}/scripts/ops-stop-hook.sh"
command -v node >/dev/null 2>&1 || exit 0; node "${CLAUDE_PLUGIN_ROOT}/scripts/ops-compress.mjs"
Self-evolving autonomous loops for Claude Code — a Stop-hook loop engine that drives an agent toward a verifiable goal with an evolving payload, checkpoint gating, and a context budget.
1 skill · 4 commands · hooks
Z.ai and OpenRouter proxy for Claude Code — stateless local router with manual /model switching and per-provider auth.
1 skill · 3 commands · hooks
Session continuity across context resets for Claude Code. Snapshots a session digest before /clear, /compact, or auto-compaction and auto-rehydrates it after. The non-looped companion to cc-repete — stands down whenever a cc-repete loop is active.
1 skill · 3 commands · hooks
A statusline framework for Claude Code that discovers per-plugin segment manifests, runs enabled ones, and joins them onto one bar — each plugin still works standalone.
2 skills · 3 commands
GLM-offload review agents for Claude Code — auto-convenes a multi-lens panel of parallel GLM reviewers on spec and plan writes and synthesizes the findings into one scored report. Requires the cc-proxy plugin to run.
Single-skill harness for surfacing unknowns before, during, and after implementation — classifies each task across known/unknown quadrants so a strong model doesn't confidently execute the wrong thing.
Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
15 skills · hooks
Persistent memory for Claude Code: the next session starts where the last one left off.
29 skills · 1 MCP server · hooks
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
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
AI memory system — mine projects and conversations into a searchable palace. 47 MCP tools, auto-save hooks, guided setup.
3 skills · 6 commands · 1 MCP server · hooks
Use Alibaba Open Code Review from local Codex.
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.