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 EZotoff/ez-omo-config --skill debategit clone --depth 1 https://github.com/EZotoff/ez-omo-configWrote 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/ezotoff/ez-omo-config/debate)<a href="https://agentmods.dev/skills/ezotoff/ez-omo-config/debate"><img src="https://agentmods.dev/badge/skills/ezotoff/ez-omo-config/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/ezotoff/ez-omo-config/debate"><img src="https://agentmods.dev/badge/skills/ezotoff/ez-omo-config/debate.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00067 | $0.07348 |
| Opus 5 | $0.00034 | $0.03674 |
| Sonnet 5 | $0.00013 | $0.01470 |
| Haiku 4.5 | $0.00007 | $0.00735 |
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 6d 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 — 775 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Debate Protocol — Structured Adversarial Analysis
Core Infrastructure
The rules below apply to every mode in this skill.
Judge Output Persistence (Persist-On-Collect)
Judge outputs are ephemeral — background_output is fire-once and compression destroys verbatim content. The filesystem is the only reliable persistence layer.
Rule: After collecting each judge's (or reviewer's) background_output, the orchestrator MUST write the raw output to disk BEFORE parsing scores, synthesizing, or doing any other work.
| Step | Action |
|---|---|
| 1 | Collect background_output(task_id="...") |
| 2 | Write raw output verbatim to judges/judge-{role}[-round-{N}].md |
| 3 | Parse scores/synthesis from the persisted file (never from memory) |
| 4 | Recompute any totals mechanically (never trust agent arithmetic) |
File path: .sisyphus/debates/{slug}/judges/judge-{role}[-round-{N}].md
Malformed output handling: If the output cannot be parsed, persist it anyway with a <!-- MALFORMED: {reason} --> comment at the top. Exclude from scoring/decision. Note in transcript.md.
Non-scoring modes (Panel Review, Writing Refinement): Judges may produce deliverables (rewrites, critiques, recommendations) instead of scores. These follow the same persist-on-collect rule. The file path is judges/judge-{role}.md (no round number).
Retry & Recovery
When a judge or reviewer agent hangs or fails, the orchestrator follows this protocol:
| Event | Action |
|---|---|
| No output after 10 minutes | background_cancel(taskId="..."), retry once with identical prompt |
| Retry also fails (10 min) | Cancel, skip this judge/reviewer for this round |
| Malformed output | Persist raw file, exclude from scoring, do not retry |
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.
- 6d ago Changed fa115c926a6a
- 10d ago First seen · 775 lines · 67 tokens per session scan A e2df93cd4fe8
debate is a skill published in the GitHub repository EZotoff/ez-omo-config (5 stars, last pushed 5d ago), licensed MIT. It adds 67 tokens to every session and 7,348 once invoked, about $0.0003 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
omc-plan
Strategic planning with optional interview workflow.
launch
Shipyard's governed delivery pipeline — converge the mission, synthesize a durable spec, decompose vertical-slice tickets with blocking edges, run the frontier in parallel via team, close with verification, and report with a full decision log. Two entry gates — the yard gate (drydock audit) and the fog gate (an effort…
ralph
Self-referential loop until task completion with configurable verification reviewer.
self-improve
Autonomous evolutionary code improvement engine with tournament selection.
autopilot
Full autonomous execution from idea to working code.
drydock
Lay the keel of the shipyard harness in any repo — the 4-pillar shared environment (Context, Rules, Tools, Standards) across 5 surfaces (CLAUDE.md, skills, design-system, mcp/cli, shared context) so that every human and agent inherits the same design language and anyone can ship. Run once per repo; re-run with --check…