ethanplusai/codex-on-crack / plugins/codex-on-crack/skills/crack
crack
UnverifiedLead substantial Codex builds with one configured implementation worker and lead acceptance. Use when the user requests codex-on-crack, names a model to delegate work to, asks to run subagents through Claude, or requests a lead/worker split. Keep host-only work direct; honor explicit external model choices. Workers must not invoke this orchestrator.
references/checkpoint.md
View on GitHub# Resumable checkpoints
A checkpoint lets the next session continue without repeating completed work. It
is a short, current status record, not an append-only transcript and not a
substitute for the repository or the tests.
## When to write one
- at a meaningful boundary: the end of a phase, before a handoff, or when you
have to stop with work unfinished;
- when a decision or constraint would be expensive to rediscover.
## What it records
- the agreed result and scope, and where the spec or plan lives;
- the workspace, baseline, and preserved user changes;
- decisions that are settled, with the reason;
- completed work and how it was verified;
- the active or unfinished work, who owns it, and its current state;
- outstanding risks, blockers, and anything unverified;
- the exact next action, and the authorization that still applies.
Keep the newest state in place of the old. Update or replace stale lines instead
of appending history. Cite file and line for important findings so the next
reader can confirm them against the repository.
The core skill includes an optional fill-in form:
[checkpoint template](../templates/checkpoint.md). In direct mode,
keep the same fields in the task notes or an equivalent local file. Keep
checkpoints local (for example under a `.crack/` directory) unless the project
wants them tracked.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |