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.
git clone --depth 1 https://github.com/hamza-ali-shahjahan/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/hamza-ali-shahjahan/claudex/stats)<a href="https://agentmods.dev/commands/hamza-ali-shahjahan/claudex/stats"><img src="https://agentmods.dev/badge/commands/hamza-ali-shahjahan/claudex/stats/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/hamza-ali-shahjahan/claudex/stats"><img src="https://agentmods.dev/badge/commands/hamza-ali-shahjahan/claudex/stats.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00016 | $0.00455 |
| Opus 5 | $0.00008 | $0.00228 |
| Sonnet 5 | $0.00003 | $0.00091 |
| Haiku 4.5 | $0.00002 | $0.00046 |
Grade A, and why
stats 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 9d 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.
This is a copy
100% identical to stats — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
What it actually says
You are running ClauDex stats — a read-only tally of the duet's history in this repository. No review happens here and no Codex call is made, so this command never signs: the 🧡🖤 line is earned by cross-review only, and counting past reviews is not one.
- Find the ledger. Confirm you are in a git repo
(
git rev-parse --git-dir); the ledger lives at<git-dir>/claudex/stats.jsonl. If the repo, the file, or its contents are missing, say so plainly: runs are recorded from ClauDex v0.6.0 onward, so older duets are invisible — "Run/claudex:verdictor/claudex <task>and come back; every run writes one line." - Tally. Parse each JSON line, skipping (and counting) malformed ones:
- Runs per command, and outcomes: signed / interrupted / deadlocked / nothing-to-review.
- The headline — agreement rate across
verdictruns:both / (both + claude_only + codex_only). - Verdict rulings: SHIP vs FIX FIRST vs REDESIGN.
- Loop runs: average rounds to consensus, and Codex's hit rate —
fixed / (fixed + rejected)(how often its findings held up). - Debates: CONSENSUS vs SPLIT.
- Interruption rate overall (the duet's reliability in this repo).
- Report. A compact table, then one headline sentence, e.g.: "Across 12 reviews the models agreed on 41% of findings — and every disagreement was a place a human looked." If there are fewer than 5 runs, say the numbers are anecdotes, not statistics, and skip the percentage framing.
- End without the sign-off line — this command neither builds, reviews, nor argues.
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.
- 9d ago First seen · 33 lines · 16 tokens per session scan A 0e8212815426
stats is a command published in the GitHub repository hamza-ali-shahjahan/claudex (4 stars, last pushed 1mo ago), licensed MIT. It adds 16 tokens to every session and 455 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. It is 100% identical to stats, differing in 0 lines, and is treated as a copy.
Other commands, from other repositories
duck-off
Turn off Rubber Duck mode and answer normally.
save
Save the current session digest. Fans out across every writable memory tier per the house-map. Tier 0 always; Tier 1 if reachable; Tier 2 only if registered writable with auth. Each tier independent — failures degrade gracefully.
check-on-claude
Maude checks on Claude — repeated tool calls, unread context, confabulation risk, missed CLAUDE.md, the patterns Claude doesn't see in himself.
notice
Maude surfaces patterns from the turn-by-turn trace — recurring topics, repeated mistakes, time-of-day patterns, sessions that keep ending in the same place.
promote
Show what the tape is holding for your word, and promote the ones you approve into canon. Agent inference never becomes canon on its own — this is the door, and your hand is on it.
help
Rubber Duck: what it does, how to steer it, how to stop it.