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 skills add donnfelker/loop-skills --skill multi-llm-convergence-betagit clone --depth 1 https://github.com/donnfelker/loop-skillsWrote 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/donnfelker/loop-skills/multi-llm-convergence-beta)<a href="https://agentmods.dev/skills/donnfelker/loop-skills/multi-llm-convergence-beta"><img src="https://agentmods.dev/badge/skills/donnfelker/loop-skills/multi-llm-convergence-beta/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/donnfelker/loop-skills/multi-llm-convergence-beta"><img src="https://agentmods.dev/badge/skills/donnfelker/loop-skills/multi-llm-convergence-beta.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 | $0.00146 | $0.02741 |
| Opus 5 | $0.00073 | $0.01371 |
| Sonnet 5 | $0.00029 | $0.00548 |
| Haiku 4.5 | $0.00015 | $0.00274 |
Grade A, and why
multi-llm-convergence-beta 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 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- Do **not** run `git clone`, `curl`, `wget`, package managers, installers, or update commands as How it starts
The opening of the file, as written. The whole thing — 210 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Multi-LLM Convergence (beta)
Beta/experimental. This is the host-agnostic, N-model variant of the stable
multi-llm-convergenceskill. Treat the stable skill as the source of truth for the loop: preflight, ground reviewers in source-of-truth, commit each round, review sequentially, and stop only on real cross-model consensus. The beta changes the dispatch layer so the same loop can be launched from Claude, Codex, Gemini, or another capable host.
You are the convergence driver. You own an artifact, and your job is to rotate it through multiple genuinely different LLM reviewer families - round after round - until every selected family independently blesses the same artifact state. You apply findings, you commit each round, and you stop only when there is real cross-model consensus or a principled stall.
Announce at start: "I'm using the multi-llm-convergence-beta skill - let me confirm the artifact, the reviewer set, and the bar, then I'll rotate the selected model families until they all agree."
Why this skill exists
A single reviewer has blind spots, and a single model family has correlated blind spots. Genuine convergence comes from alternating different model families and letting each catch what another introduced or missed.
The stable skill proves the loop with two reviewer families. This beta preserves that methodology and generalizes only the reviewer dispatch:
- Host-agnostic launch. The host can be Claude, Codex, Gemini, or another environment that can run the selected official reviewer CLIs. The host model is the orchestrator, not an implicit reviewer.
- N-model reviewer set. The operator may select any two or more built-in reviewer families with
distinct
familynames. Built-ins are documented inreferences/reviewer-profiles.md. - One reviewer protocol. Every reviewer uses the same lifecycle: fixed built-in profile, smoke-test, read-only mode, identical review contract, captured output, liveness supervision, structured JSON parsing, then the same apply-and-commit step.
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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 · 210 lines · 146 tokens per session scan A a19579e898b6
multi-llm-convergence-beta is a skill published in the GitHub repository donnfelker/loop-skills (18 stars, last pushed 1mo ago), licensed MIT. It adds 146 tokens to every session and 2,741 once invoked, about $0.0007 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…