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 wjgoarxiv/autoresearch-skill --skill reasongit clone --depth 1 https://github.com/wjgoarxiv/autoresearch-skillWrote 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/wjgoarxiv/autoresearch-skill/reason)<a href="https://agentmods.dev/skills/wjgoarxiv/autoresearch-skill/reason"><img src="https://agentmods.dev/badge/skills/wjgoarxiv/autoresearch-skill/reason/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/wjgoarxiv/autoresearch-skill/reason"><img src="https://agentmods.dev/badge/skills/wjgoarxiv/autoresearch-skill/reason.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 | $0.00123 | $0.01492 |
| Opus 5 | $0.00062 | $0.00746 |
| Sonnet 5 | $0.00025 | $0.00298 |
| Haiku 4.5 | $0.00012 | $0.00149 |
Grade A, and why
autoresearch:reason 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 10d 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 — 163 lines — stays where its author put it; the contents beside it link to each section on GitHub.
autoresearch:reason
Adversarial multi-round reasoning loop with a blind-judge panel. Arguments are assigned crypto-random IDs before critique so judges evaluate logic, not author identity. Runs until convergence or budget exhausted.
Autonomy Directive
You are an autonomous reasoning agent. Once the debate begins:
- NEVER STOP to ask permission between rounds.
- NEVER ASK "should I continue?" mid-debate.
- NEVER DECLARE CONVERGENCE prematurely — a single unchallenged round is not convergence.
- The loop runs until: all positions converge OR budget exhausted OR user interrupts.
- If neither condition is true, begin the next round immediately.
Setup
Step 1 — Define the question
If not provided, ask once: "What is the question or decision to reason about?" The question must be specific enough to allow falsifiable positions. Refuse vague inputs like "think about AI" — ask for a concrete framing.
Step 2 — Set parameters
Ask if not provided:
- Number of positions (default: 3). Minimum 2, maximum 5.
- Max rounds (default: 4). Minimum 2.
- Convergence threshold: "Converged" = all judges rate the top position's logical score ≥8/10 AND no position has an unanswered rebuttal.
Round Structure
Each round follows this exact sequence:
Phase 1 — Propose Positions (Round 1 only)
Generate N distinct positions on the question. Positions must:
- Be mutually distinguishable (not minor variations of each other)
- Be stated as falsifiable claims, not vague preferences
- Cover the genuine range of defensible views (not strawmen)
Write each position to reason/rounds.md as: Position [PENDING-ID]: [statement]
Phase 2 — Assign Crypto-Random IDs
Assign each position a random alphanumeric ID (e.g., ARG-7F3A, ARG-2C91).
Write the mapping to reason/id-map.md — this file is sealed until the end (not read during debate).
Replace all position labels in reason/rounds.md with their assigned IDs.
From this point forward, all debate references use IDs only — never "Position 1" or author names.
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.
- 10d ago First seen · 163 lines · 123 tokens per session scan A f91e7e85060f
autoresearch:reason is a skill published in the GitHub repository wjgoarxiv/autoresearch-skill (32 stars, last pushed 2mo ago), licensed MIT. It adds 123 tokens to every session and 1,492 once invoked, about $0.0006 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 2026-08-30.
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…