cccc

A local handoff tool that sends one bounded coding request to another installed Claude Code or Codex command-line agent. It can ask that agent to implement a scoped change or give a read-only second opinion.

In plain words
What is it for?
Use it to delegate a small implementation or request another agent’s review of code and decisions.
Why use it?
It lets you use a second local coding agent without losing control of the task or accepting unverified results.

Skill for Claude CodeCodex

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/jtropy/cccc-skill/cccc
Any agent
npx skills add JTropy/cccc-skill --skill cccc
Clone the repo
git clone --depth 1 https://github.com/JTropy/cccc-skill

Made for: Claude Code, Codex.

Per session 58 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 682 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00058 $0.00682
Opus 5 $0.00029 $0.00341
Sonnet 5 $0.00012 $0.00136
Haiku 4.5 $0.00006 $0.00068

Measured 2d ago against content hash 1fe8dbfe3f94, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

cccc scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

The scan reads SKILL.md. This mod also ships 6 executable files (scripts/cccc-common.sh, scripts/consult.sh, scripts/delegate.sh, …), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/cccc/SKILL.md · 58 lines

How it starts

The opening of the file, as written. The whole thing — 58 lines — stays where its author put it; the contents beside it link to each section on GitHub.

cccc

Route one bounded request to the other installed local CLI. delegate permits scoped implementation; consult requests a read-only second opinion. The orchestrator validates the result and makes no unverified capability claim.

1. Authorization gate

  • An explicit request naming cccc/4C, this wrapper, or a handoff to the installed local Claude Code or Codex peer authorizes that path within the user's scope.
  • Bare delegate or consult terms are not launch authorization; treat them as a heuristic until the user confirms the local peer handoff.
  • For a heuristic match, only suggest cccc. Confirm with the user before any external CLI launch.
  • If user or project instructions prohibit external agents, do not launch a wrapper.

Discovery of this skill is not launch authorization. Apply this gate again immediately before invoking a wrapper.

2. Routing precedence

Use this order exactly:

user instruction > project rules > verified local capability > heuristic

The Codex orchestrator targets Claude; the Claude Code orchestrator targets Codex. Never target itself or infer identity from executable availability.

3. Choose one channel

Need Channel Result
Bounded repository implementation with explicit editable paths delegate Peer changes plus a report and audit log
Analysis, critique, or decision support without repository edits consult Peer opinion plus an audit log

Do not use consult as authorization to implement its recommendation. Do not use delegate when the user asked only for analysis.

4. Verify capability

Verify the selected CLI and authentication before launch. Follow setup when a check fails.

For raster images, verify usable Codex $imagegen capability first. Do not promise image generation, a model, effort, cost, or network path.

5. Safety invariants

  • Use only the canonical wrappers in this skill. Never construct a looser raw CLI fallback.
  • Treat wrapper exit status 0 as the success authority; an existing report or opinion alone is not success.
  • DELEGATE_DEPTH and wrapper prompts guard against accidental peer recursion; do not ask the peer to invoke another agent.
  • Tool restrictions limit exposed operations. The post-run Git audit detects Git-visible changes; neither control is an OS sandbox. Use external isolation for an untrusted repository.
  • The peer must not alter repository history, hide working state, discard files, publish changes, or broaden the requested scope.

Read the full file on GitHub · 58 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 58 lines · 58 tokens per session scan A 1fe8dbfe3f94

Subscribe to this mod's changes

cccc is a skill published in the GitHub repository JTropy/cccc-skill (10 stars, last pushed 20d ago), licensed MIT. It adds 58 tokens to every session and 682 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

brainstorming

You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.

obra/superpowers · 37 tokens

chat-pet-sprite-creation

Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.

microsoft/vscode · 53 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens

agent-host-chat-contributions

Build and review cross-cutting agent-host chat behavior through lifecycle contributions. Use when adding turn lifecycle side effects, prompt or context injection, restored-history transformation, protocol-action observation, or when reviewing changes that add code to AgentSideEffects or AgentService.

microsoft/vscode · 56 tokens

auto-perf-optimize

Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.

microsoft/vscode · 62 tokens