Borrowing it
Nothing to install: this file belongs to adrienm7/ergopti. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/adrienm7/ergopti/main/.agents/skills/orchestrate-pass/SKILL.mdgit clone --depth 1 https://github.com/adrienm7/ergoptiWrote 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/skills/adrienm7/ergopti/orchestrate-pass)<a href="https://agentmods.dev/skills/adrienm7/ergopti/orchestrate-pass"><img src="https://agentmods.dev/badge/skills/adrienm7/ergopti/orchestrate-pass/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/skills/adrienm7/ergopti/orchestrate-pass"><img src="https://agentmods.dev/badge/skills/adrienm7/ergopti/orchestrate-pass.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 | $0.00027 | $0.00862 |
| Opus 5.5 | $0.00011 | $0.00345 |
| Sonnet 5.5 | $0.00005 | $0.00172 |
| Haiku 4.5 | $0.00003 | $0.00086 |
Grade A, and why
orchestrate-pass 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 15d 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 — 84 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Running a multi-agent pass
Fan-out earns its cost on breadth — many files, many independent questions. It does not earn it on depth, and it never earns it on writing.
Scout first, then fan out
Discover the work list in your own context (list the files, find the sites, scope the diff), then parallelise over it. Fanning out before you know the shape produces agents that each rediscover the same landscape.
Read-only fan-out, sequential writes
Agents research and verify. You implement, one commit at a time. Parallel
agents editing the same tree conflict on shared files — tests/run_all.ahk is the
obvious one, since every new test must be registered there.
Feed them what has already been rejected
The single highest-value thing you put in an agent prompt. Without it they will
confidently re-propose ideas this repo already tried and reverted — tooltip window
reuse, chunking the emoji registration, process-priority changes. Extract the
rejected list from docs/memory/rejected_proposals.md and paste it in with the
reasons.
Also give them: the real paths (including the config redirect, see
perf-profiling), the conventions that make a proposal undeliverable, and the
evidence standard — every claim carries file:line, and every number is labelled
measured or deduced.
Verify with two independent lenses
One refuter is a coin flip. Two agents looking from different angles is what catches a plausible-but-wrong finding:
- Code lens — does the cited code actually do this? Recount the per-keystroke cost yourself. Is the transformation correct in this language?
- Memory and convention lens — does this re-raise something rejected? Is the cache invalidation provable? Does the named test exist, and does it cover the invariant?
One refutation kills the finding. In practice this rejects roughly half, which is the point.
A refutation is a hypothesis too. Verify the verifier when the stakes are high: a lens once dismissed a measured 38 ms parse as "implausible, realistically 2-5 ms". Re-benching gave 44.3 ms — the finding was right and the refutation was wrong. Cheap to check, expensive to get backwards.
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.
- 15d ago First seen · 84 lines · 27 tokens per session scan A 2b2e414e110f
orchestrate-pass is a skill published in the GitHub repository adrienm7/ergopti (14 stars, last pushed yesterday), licensed MIT. It adds 27 tokens to every session and 862 once invoked, about $0.0001 per session on Opus 5.5. 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 2026-10-14.
Other skills, from other repositories
distributed-training-deepspeed
Expert guidance for distributed training with DeepSpeed - ZeRO optimization stages, pipeline parallelism, FP16/BF16/FP8, 1-bit Adam, sparse attention.
ml-training-recipes
Battle-tested PyTorch training recipes for all domains — LLMs, vision, diffusion, medical imaging, protein/drug discovery, spatial omics, genomics. Covers training loops, optimizer selection (AdamW, Muon), LR scheduling, mixed precision, debugging, and systematic experimentation. Use when training or fine-tuning…
cuopt-skill-evolution
After solving a non-trivial problem, detect generalizable learnings and propose skill updates. Always active — applies to every interaction.
seedance-vocab-en
This skill should be used when an English Seedance 2.0 prompt needs clearer production wording, less generic prose, or precise vocabulary for camera, lighting, motion, VFX, audio, and constraints. Route blocked prompts through seedance-filter for context and boundary review.
prompt-optimization
Use this skill when the user wants to optimize, modify, or improve the system prompt of an AI agent. This includes requests like 'optimize the prompt', 'make the AI more focused on X', 'change the system prompt', 'improve the agent behavior', or 'modify how the AI responds'.
seo-audit
Audit affiliate blog posts and landing pages for SEO issues. Triggers on: "audit my blog post for SEO", "check my SEO", "SEO review", "improve my rankings", "SEO checklist", "on-page SEO audit", "keyword optimization check", "why isn't my page ranking", "SEO score", "content quality audit", "check my meta tags"…