From pulp
Invokes OpenAI Codex via stdin to get a second opinion, handling argv-hang and timeout gotchas. Reads prompt from $ARGUMENTS or a file.
How this command is triggered — by the user, by Claude, or both
Slash command
/pulp:codex-consultcommands/The summary Claude sees in its command listing — used to decide when to auto-load this command
# Codex Consult A Pulp-side wrapper around `codex exec` that always invokes it the right way. Use this instead of calling `codex exec` ad hoc — the direct CLI silently hangs on anything past a short argv prompt. ## When to use - You want a second opinion on a plan, review, or design choice - You need to sanity-check assumptions about a tool/API before committing to a direction - You want Codex to explore the repo independently (it can use the RepoPrompt MCP server from inside its session) See `~/.claude/skills/codex/SKILL.md` for the full flag reference and gotchas. ## How to invoke A...
A Pulp-side wrapper around codex exec that always invokes it the right way. Use this instead of calling codex exec ad hoc — the direct CLI silently hangs on anything past a short argv prompt.
See ~/.claude/skills/codex/SKILL.md for the full flag reference and gotchas.
Agents: build the prompt in a file, then run the invocation below. Never pass a long prompt as a single argv argument to codex exec — it hangs silently with 0% CPU.
# 1. Write the prompt to a temp file (heredoc into `cat > /tmp/codex-prompt.txt <<'EOF' … EOF`).
# 2. Invoke:
cat /tmp/codex-prompt.txt | timeout 480 \
codex exec -c model_reasoning_effort='"medium"' - \
> /tmp/codex-answer.txt 2> /tmp/codex-progress.log
# 3. Read the answer:
cat /tmp/codex-answer.txt
Tune the reasoning effort down ("low") for faster answers on simple questions, up ("high", "xhigh") for deep analysis — but expect multi-minute runs at xhigh.
Include this framing in every Codex prompt so it knows to use RepoPrompt instead of raw shell, and knows the repo state:
Tooling preference: use the RepoPrompt MCP server's tools (file_search,
get_file_tree, read_file, get_code_structure) rather than raw shell.
CWD is /Users/danielraffel/Code/pulp on branch <BRANCH> at commit <SHA>.
<your question here>
Please give direct yes/no answers where possible and flag any step in
the plan that is wrong or risky.
If the background invocation isn't returning:
ps -p <codex_pid> -o pid,state,time,%cpu,rss
lsof -p <codex_pid> 2>/dev/null | grep TCP | head -3
tail -20 /tmp/codex-progress.log
mcp: …started / …completed lines.codex exec "long prompt" as a single argv argument — always pipe via stdin (codex exec -).timeout — a hung codex is indistinguishable from a slow one.npx claudepluginhub danielraffel/pulp --plugin pulp/consult-codexConsults OpenAI Codex CLI on current repo for code analysis, review, and refactoring; synthesizes with Claude's reasoning in 5-part response.
/codexSends a single question to Codex (OpenAI's coding model) and relays its answer, providing a quick second opinion alongside Claude. Includes privacy check and user confirmation before sending.
/codex-peer-reviewTriggers peer review validation using OpenAI Codex CLI. Defaults to symmetric blind-pass debate on git changes. Supports branches, commits, uncommitted code, profile init, classic mode, and questions.
/codexDelegates coding tasks to OpenAI Codex CLI for autonomous execution via natural language prompts, handling code refactoring, testing, bug fixes, explanations, features, and reviews.
/codex-reviewOrchestrates a deep code review using Codex jobs for detection, validation, and cross-cutting analysis, producing an artifact.json compatible with the :review workflow.
/codex-callBuilds and executes a Codex CLI call via the runner, handling prompt assembly (persona, provenance, conventions), fresh/resume semantics, timeouts, heartbeats, and job registration.