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 nyxandro/sparring-skill --skill sparringgit clone --depth 1 https://github.com/nyxandro/sparring-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/nyxandro/sparring-skill/sparring)<a href="https://agentmods.dev/skills/nyxandro/sparring-skill/sparring"><img src="https://agentmods.dev/badge/skills/nyxandro/sparring-skill/sparring/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/nyxandro/sparring-skill/sparring"><img src="https://agentmods.dev/badge/skills/nyxandro/sparring-skill/sparring.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 | $0.00154 | $0.03175 |
| Opus 5.5 | $0.00062 | $0.01270 |
| Sonnet 5.5 | $0.00031 | $0.00635 |
| Haiku 4.5 | $0.00015 | $0.00317 |
Grade A, and why
sparring 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 today.
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 — 215 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Sparring
Use this skill after it has been selected for a sparring request with another CLI agent. The skill turns that request into a concrete workflow through the bundled sparring harness.
Production rule: use ask-session with a unique temporary state file. The same command creates the provider-native session on the first turn and resumes it on later turns. Use tmux only when non-interactive CLI mode is unavailable or when the user explicitly wants a live attachable TUI session; do not expose tmux as a second normal workflow.
Scope
After this skill is selected, use it only for sparring-style requests with another model or CLI agent. Do not use it for ordinary non-interactive shell commands where stdout alone is the requested result.
Requirements
This skill expects:
- The bundled
bin/sparctl,lib/print-agent.sh, andlib/tmux-agent.shfiles from this skill folder. - The requested opponent CLI installed, for example
claudeorcodex. timeoutfrom GNU coreutils for non-interactive CLI calls.tmuxonly when using explicit live TUI mode.
Resolve the harness path in this order:
- Use this skill's own bundled harness:
<skill directory>/bin/sparctl. - If the runtime exposes a “Base directory for this skill” line, treat that directory as
<skill directory>. - If you are developing the skill locally and
SPARRING_HOMEis explicitly set, it may override the bundled path for testing only. - If no executable bundled
bin/sparctlexists, reportSPARRING_HARNESS_MISSINGand include the skill directory you checked.
Do not silently fall back to raw tmux when sparctl is missing; the harness contains the tested session and tmux behavior.
Language Rule
Use the user's request language for the opponent prompt and for the final response.
If the user writes in Russian, send the opponent a Russian prompt and return Russian output. If the user writes in English, send the opponent an English prompt and return English output. If the user mixes languages, prefer the dominant language of the task.
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.
- today Changed · +2 lines 816ae1303a46
- 2mo ago First seen · 213 lines · 154 tokens per session scan A 874b52d23e27
sparring is a skill published in the GitHub repository nyxandro/sparring-skill (3 stars, last pushed yesterday), licensed MIT. It adds 154 tokens to every session and 3,175 once invoked, about $0.0006 per session on Opus 5.5. 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.
sparring compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| sparringnyxandro/sparring-skill | Skill | 3 | yesterday | MIT | A | 154 | Claude Code, Codex |
| socratic-reviewthoughtbot/rails-consultant | Skill | 28 | 7d ago | MIT | A | 51 | Claude Code |
| prior-artthoughtbot/rails-consultant | Skill | 28 | 7d ago | MIT | A | 36 | Claude Code |
| paired-sessionNYTC69/review-loop | Skill | 3 | 2d ago | Apache-2.0 | A | 0 | Claude Code, Codex |
| ship-it-or-fix-itEzra144israel/governed-agent-skills | Skill | 22 | 1mo ago | MIT | A | 98 | Claude Code |
| test-verificationEzra144israel/governed-agent-skills | Skill | 22 | 1mo ago | MIT | A | 43 | Claude Code |
| reorganizeNYTC69/review-loop | Skill | 3 | 2d ago | Apache-2.0 | A | 32 | Claude Code |