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/runnithan/claudex-setup/optimisegit clone --depth 1 https://github.com/runnithan/claudex-setupWhat 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 | $0.00038 | $0.09816 |
| Opus 5 | $0.00019 | $0.04908 |
| Sonnet 5 | $0.00008 | $0.01963 |
| Haiku 4.5 | $0.00004 | $0.00982 |
Grade A, and why
optimise scanned grade A with 1 finding 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 2d 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.
Reads agent configuration directorieslowAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
`AGENTS.md`, any project `.codex/` directory, and the relevant `~/.codex/config.toml`, the Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.
How it starts
The opening of the file, as written. The whole thing — 622 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/optimise: audit a project's Claude setup and apply improvements
You are auditing ONE registered project's Claude tooling setup against the repo's
lessons/ library, then walking the owner through concrete improvements interactively and
applying the ones they approve. This is not a code review of the project's application
code, every finding is about how the owner works with Claude on this project (config,
agents, skills, hooks, prompting/keyboard habits), never about whether the app itself is
buggy.
This command runs from inside the claudex-setup repo (the hub). The project being audited
usually lives elsewhere on disk (its real path is in the registry frontmatter). You edit two
places: the audited project's actual .claude/ (applying approved config changes), and the
claudex-setup repo's projects/ tracking files (recording what you found and did).
Tune your behaviour to run reliably and unattended-friendly: follow the numbered steps in order, use the exact file paths and formats below, stop where the stop conditions say to stop, and run the final self-check before reporting done.
0. Inputs and mode resolution
- Read the argument string. The first bare token is the project slug (e.g.
your-project,claudex-setup,another-project,personal-notes). If no slug is given, list the registered projects (the subdirectories ofprojects/, excluding the root-level registry filesREADME.md,audit-log.md, andhabits.md) and ask the owner which one to audit, then stop until they answer. - Detect a tool flag anywhere in the argument string:
--designselects the Claude Design sub-area,--codexselects the Codex sub-area, and--code(or no flag) selects the default Claude Code area. SetMODE = design,MODE = codex, orMODE = codeaccordingly. If the project is multi-tool and no flag was given, tell the owner the project has sub-areas and which flags reach the others (--design,--codex), the default must be a visible choice, not a silent one, then proceed withclaude-code/.--designor--codexon a flat (single-tool) project is an error: only a multi-tool project has those sub-areas. - Establish the repo root. This command lives in the claudex-setup repo; the registry is at
<repo-root>/projects/. Use<repo-root>as the base for everyprojects/...path below.
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.
- 2d ago First seen · 622 lines · 38 tokens per session scan A 115f08453082
optimise is a command published in the GitHub repository runnithan/claudex-setup (1 stars, last pushed 2d ago), licensed MIT. It adds 38 tokens to every session and 9,816 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
multi-agent-optimize
The Multi-Agent Optimization Tool is an advanced AI-driven framework designed to holistically improve system performance through intelligent, coordinated agent-based optimization. Leveraging cutting-edge AI orchestration techniques, this tool provides a comprehensive approach to performance engineering across multiple…
ship-an-mcp-server
Workflow recipe — make your product agent-usable by chaining 4 skills, spec to pricing.
auto
Intent-classified router, the front door to OrchestKit and the DEFAULT entry point for any goal-shaped request. Classifies a plain-English goal and routes it to the right specialist skill. Routing is never overhead, so use it even when the target skill seems obvious; skip only when already executing inside another…
loop-forge
Turn a one-line repetitive task into a reusable, self-guarding slash command (loop-forge).
pack-repo
Pack la codebase en un fichier AI-friendly (Repomix wrapper + fallback shell). Token counting inclus.
van
Command "van" from vanzan01/claude-code-sub-agent-collective, covering /van - collective routing engine, 🎯 purpose - smart routing, 🚐 routing flow, 🚀 dual-mode routing protocol and 🧠 immediate agent routing.