runnithan/claudex-setup

runnithan/claudex-setup

Command Claude Code

Cancel an active Ralph Loop.

runnithan/claudex-setup

Command Claude Code

Use Codex as a second, differently-trained reviewer of your branch, fix every real finding with regression tests, and loop until two consecutive reviews come back clean. Runs autonomously.

runnithan/claudex-setup

Command Claude Code

Consolidate the lessons/ archive: merge near-duplicate lessons, split compound ones, prune superseded files, refresh INDEX. Counterpart to /extract-lessons.

runnithan/claudex-setup

Command Claude Code

Extract actionable agent-tooling lessons (Claude Code, Claude Design, Codex) from transcripts. Newer lessons supersede older contradicting ones.

runnithan/claudex-setup

Command Claude Code

Mine Claude Code lessons from recent posts by notable practitioners (X/Twitter, blogs, GitHub) and canonical sources (changelog, docs, issues, Hacker News) via a research subagent team. Counterpart to /extract-lessons (transcripts).

runnithan/claudex-setup

Command Claude Code

Verify claims in content by searching for sources. Use when user says /fact-check followed by content to verify.

runnithan/claudex-setup

Command Claude Code

Audit a tracked project's Claude Code, Claude Design, or Codex setup, fold in new lessons, then apply open improvements one at a time with approval. Takes a project name from projects/.

runnithan/claudex-setup

Command Claude Code

Create a PR from current branch targeting main, with Jira ticket reference.

runnithan/claudex-setup

Command Claude Code

Review an agent-written change by being examined on it: quizzes you on the diff, diagrams what moved, and turns the questions you cannot answer into the review findings. Use after a large agent-produced change instead of reading every line.

runnithan/claudex-setup

Command Claude Code

Explain Ralph Loop plugin and available commands.

runnithan/claudex-setup

Command Claude Code

Start a Ralph Loop, iterative self-referential development loop.

runnithan/claudex-setup

Command Claude Code

Comprehensive PR review using specialized agents.

runnithan/claudex-setup

Command Claude Code

Update CLAUDE.md with learnings from this session.

runnithan/claudex-setup

Command Claude Code

Run the project's tests, lint, and build checks. Optionally scope to one suite, e.g. /test backend.

runnithan/claudex-setup

Command Claude Code

Pick up a Jira ticket: fetch details, transition to In Progress, create branch, checkout locally.

runnithan/claudex-setup

Command Claude Code

Rank the high-value lessons that never reached the owner (no trace in habits, config, or backlogs) and offer the top few for adoption. Complements /optimise: it delivers the class with no config surface.

runnithan/claudex-setup

Command

Prefer giving agents small CLI tools over MCPs: a CLI is read once then driven by command, is deterministic, self-documents via --help, and costs far less context than an MCP's loaded tool schemas.

runnithan/claudex-setup

Command

Package the "ask before you build" habit into a reusable custom slash command: it captures the task, asks 5 clarifying questions, drafts a plan, and waits for approval before executing.

runnithan/claudex-setup

Command

/effort sets how deeply the model reasons: low is fast and shallow, the default is high, and extra-high/max are slower but markedly more careful, switch up for risky or complex changes, back down for routine ones, and check the cost with /usage afterwards.

runnithan/claudex-setup

Command

The environment selector at the bottom of Claude Code switches where work executes: local for fast interactive iteration, cloud (Anthropic servers) for routines that must run with your machine off, SSH for driving a remote box like a VPS.

runnithan/claudex-setup

Command

TL;DR: Embed bash in a slash command to pre-compute state (e.g. git status) so the model gets the context for free instead of round-tripping for it.