Borrowing it
Nothing to install: this file belongs to doncheli/don-cheli-sdd. 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/doncheli/don-cheli-sdd/main/.agent/skills/doncheli-debate/SKILL.mdgit clone --depth 1 https://github.com/doncheli/don-cheli-sddWrote 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/doncheli/don-cheli-sdd/doncheli-debate)<a href="https://agentmods.dev/skills/doncheli/don-cheli-sdd/doncheli-debate"><img src="https://agentmods.dev/badge/skills/doncheli/don-cheli-sdd/doncheli-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/doncheli/don-cheli-sdd/doncheli-debate"><img src="https://agentmods.dev/badge/skills/doncheli/don-cheli-sdd/doncheli-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 | $0.00056 | $0.00486 |
| Opus 5 | $0.00028 | $0.00243 |
| Sonnet 5 | $0.00011 | $0.00097 |
| Haiku 4.5 | $0.00006 | $0.00049 |
Grade A, and why
doncheli-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 11d 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.
What it actually says
Don Cheli: Adversarial Debate Panel
Roles
Each role is played by a separate internal agent with its own agenda:
| Role | Primary Concern |
|---|---|
| CPO | User value, market fit, time-to-market |
| Architect | Scalability, maintainability, tech debt |
| QA Lead | Reliability, testability, edge cases |
| UX Lead | Usability, accessibility, consistency |
| Business | Cost, ROI, risk, compliance |
Instructions
- Read the decision or proposal to evaluate
- Each role presents its position (2–4 bullets) — what it likes and what it fears
- Each role MUST identify at least one concrete problem with every other role's proposal (no free passes)
- After all attacks, each role proposes one mitigation or compromise
- Synthesize tensions and produce a decision with explicit trade-offs accepted
Output Format
## Debate: <topic>
### Round 1 — Positions
**CPO:** …
**Architect:** …
**QA Lead:** …
**UX Lead:** …
**Business:** …
### Round 2 — Attacks (each role vs. others)
**CPO → Architect:** …
**Architect → CPO:** …
… (all cross-attacks)
### Round 3 — Mitigations
**CPO proposes:** …
… (one per role)
### Decision
**Chosen approach:** …
**Trade-offs accepted:** …
**Conditions / guardrails:** …
Quality Gate
- No role may agree unconditionally with another — every role must raise at least one concern
- The Decision section must name at least two trade-offs explicitly accepted
- If consensus is impossible, output "DEADLOCK" and list the 2–3 unresolved tensions for human escalation
Do not use this skill when
- The decision is already made and the user only wants implementation help
- The question has a single objectively correct answer (use doncheli-reasoning instead)
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.
- 11d ago First seen · 65 lines · 56 tokens per session scan A f1ecbeec92d3
doncheli-debate is a skill published in the GitHub repository doncheli/don-cheli-sdd (57 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 56 tokens to every session and 486 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 2026-08-30.
Other skills, from other repositories
autonomous
Run multiple phases hands-free. Chains discuss, plan, build, and verify automatically.
quick
Execute an ad-hoc task with atomic commits. Skips full plan/review.
audit
Review past Claude Code sessions for PBR workflow compliance and UX quality.
debug
Systematic debugging with hypothesis testing. Persistent across sessions.
explore
Explore ideas, think through approaches, and route insights to the right artifacts.
resume
Pick up where you left off. Restores context and suggests next action.