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.
npx agentmods add commands/promptadvisers/claudex/plangit clone --depth 1 https://github.com/promptadvisers/claudexWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/promptadvisers/claudex/plan)<a href="https://agentmods.dev/commands/promptadvisers/claudex/plan"><img src="https://agentmods.dev/badge/commands/promptadvisers/claudex/plan.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00031 | $0.01724 |
| Opus 5 | $0.00015 | $0.00862 |
| Sonnet 5 | $0.00006 | $0.00345 |
| Haiku 4.5 | $0.00003 | $0.00172 |
Grade A, and why
plan 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 3d ago.
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.
How it starts
The opening of the file, as written. The whole thing — 122 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/claudex:plan
User argument: $ARGUMENTS
You are running the claudex plan-mode autonomous loop. The user wants you to draft a plan, hand it to Codex for adversarial review, revise based on findings, and loop until Codex has nothing material left to flag (or until the round budget is exhausted).
Optional flags
Parse these flags from the start of $ARGUMENTS (the script handles them; you mainly need to detect --skip-interview):
--rounds N. Override the default max rounds (3). Common picks: 3 (default, fast), 5 (deeper grilling), 7+ (very high stakes).--from-draft. Use the existingPLAN.mdin the project root instead of drafting from scratch. PLAN.md must exist and be non-empty.--skip-interview. Bypass the topic-sharpening interview offer in step 2 below. Useful when you've already nailed the topic or you're in a rush.
Procedure
1. Empty arguments → ask for a topic
If $ARGUMENTS is empty (no topic, no flags), do NOT run start-loop.sh. Instead, ask the user a single question:
What feature or change should I plan? (e.g. "add expiry dates to my links", "migrate auth to Clerk")
Optional: prefix with
--rounds Nto override the default of 3 grilling rounds,--from-draftto use an existing PLAN.md, or--skip-interviewto skip the interview offer.
Wait for their reply, then re-invoke /claudex:plan <their answer>. Do not proceed past this step until you have a topic.
2. Offer the topic-sharpening interview
If $ARGUMENTS does NOT contain --skip-interview, use the AskUserQuestion tool to offer the user a quick interview before launching the loop. The interview lets the user front-load constraints and worries that would otherwise come out only when Codex flags them on round 2 or 3, a much cheaper signal capture.
Use this exact AskUserQuestion call:
- Question: "Want me to interview you to sharpen the topic before launching the adversarial loop, or just go?"
- Header: "Interview"
- Options:
- Label: "Interview me first (Recommended)", Description: "I'll ask 3 quick questions about scope, constraints, and edge cases you're worried about. Then I launch the loop with a sharper topic. Costs ~30 seconds, saves a round of grilling."
- Label: "Just launch the loop", Description: "Skip the interview and start drafting PLAN.md from the topic as written. Faster, but Codex may surface basic gaps you could have answered up front."
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.
- 3d ago First seen · 122 lines · 31 tokens per session scan A 598e8df4ef2d
plan is a command published in the GitHub repository promptadvisers/claudex (119 stars, last pushed 4mo ago), licensed MIT. It adds 31 tokens to every session and 1,724 once invoked, about $0.0002 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 2026-09-01.
Other commands, from other repositories
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.