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 sk-lim19f/Hypomnema --skill debategit clone --depth 1 https://github.com/sk-lim19f/HypomnemaWrote 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/sk-lim19f/hypomnema/debate)<a href="https://agentmods.dev/skills/sk-lim19f/hypomnema/debate"><img src="https://agentmods.dev/badge/skills/sk-lim19f/hypomnema/debate/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/sk-lim19f/hypomnema/debate"><img src="https://agentmods.dev/badge/skills/sk-lim19f/hypomnema/debate.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 | $0.01194 |
| Opus 5 | $0.00062 | $0.00597 |
| Sonnet 5 | $0.00025 | $0.00239 |
| Haiku 4.5 | $0.00012 | $0.00119 |
Grade A, and why
debate 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 8d 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 — 89 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are running /hypo:debate. It formalizes an informal practice (a two-worker adversarial
review) into a reproducible fixed order, so the same rigor applies every time.
The premise: separate the mind that generates and interrogates from the mind that verifies. Interrogation without verification lets a plausible wrong answer survive; verification without interrogation confirms a wrong premise with precision. The order is fixed for that reason.
The debate always ends in a durable wiki artifact: a refreshed page, a correction, or an ADR. A debate that leaves nothing on disk was just a conversation. This is the producer reviewer pattern applied to a single claim; for the orchestration vocabulary it belongs to, see references/orchestration-patterns.md.
Two uses
- Re-verify a wiki claim (primary).
/hypo:verifysurfaces pages whoseverify_by_dateis overdue (or that carry noverify_byquestion at all). This skill is the deep re-check behind that prompt: does the page'sverify_byquestion still hold against the current code and docs? The outcome updates the page exactly as/hypo:verifyStep 4 does (see below), with the debate supplying the evidence. - Harden a hard-to-reverse decision. Architecture choices, refactor direction, external API or dependency integration, where the first-pass answer is often wrong (concurrency, migrations, security boundaries, distributed state). The decision must land as an ADR or a wiki page; if it would not, it is not big enough for a debate.
When not to use
- A change whose diff you can state in one sentence. Just do it.
- A reversible call, an already-obvious choice, or a one-off lookup. The debate's fixed cost exceeds its value.
Fixed three phases
Phase 1. Interrogate (do not answer yet)
Attack the claim or decision. Build a list of questions, not conclusions.
- Surface hidden assumptions: "this rests on X being true. What if X is false?"
- Stand up at least two alternatives (three including the incumbent). Best and worst case of each.
- List failure modes: how does this decision bite six months from now?
- Output: a list of falsifiable claims. Each must be resolvable to true / false by code,
docs, or a command ("p95 under 50ms", not "it will be fast"). For a wiki claim, restate the
page's
verify_byquestion in a form you can actually check against the current code or docs.
What ships with it
1 file 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.
- 8d ago First seen · 89 lines · 123 tokens per session scan A 19d2c807000b
debate is a skill published in the GitHub repository sk-lim19f/Hypomnema (10 stars, last pushed yesterday), licensed MIT. It adds 123 tokens to every session and 1,194 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 2026-08-31.
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…
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…
chronicle
Analyze Copilot session history for standup reports, usage tips, session search, and session reindexing. Use when the user asks for a standup, daily summary, usage tips, workflow recommendations, wants to search or find past sessions by keyword/file/PR, wants to reindex their session store, or asks about deleting…