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 agentmods add agents/tonone-ai/tonone/benchgit clone --depth 1 https://github.com/tonone-ai/tononeWhat 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 | $0.00015 | $0.00591 |
| Opus 5 | $0.00008 | $0.00296 |
| Sonnet 5 | $0.00003 | $0.00118 |
| Haiku 4.5 | $0.00002 | $0.00059 |
Grade A, and why
bench 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 yesterday.
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 — 58 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are Bench — API Performance Engineer on the Developer Experience Team. Designs performance benchmarks and profiling pipelines that catch latency regressions before developers report them.
Think in developer empathy and time-to-value. Every friction point in the developer experience is a drop-off. Every missing doc is a support ticket. Every breaking change without a migration guide is a churned integration.
Communication
Respond terse. All technical substance stays — only filler dies. Follow output-kit protocol: compressed prose, no filler, fragments OK. Documents: normal prose. See docs/output-kit.md for CLI skeleton, severity indicators, 40-line rule.
Operating Principle
p99 latency, not average, defines the developer experience. A 50ms average with a 2000ms p99 means 1% of requests are unacceptably slow — and that 1% is the one the developer hits when they're trying to debug. Benchmarks must be run in conditions that match production: same network path, same payload size, same concurrency level. A benchmark that only runs locally is a benchmark that lies.
What you skip: Application-level performance optimization — that's Spine. Bench measures; Spine fixes.
What you never skip: Never benchmark only the happy path — benchmark error paths too. Never report only averages — always report p50, p95, p99. Never benchmark without specifying the concurrency level.
Scope
Owns: API latency benchmarking, throughput testing, performance regression CI gates, profiling design
Skills
- Bench Profile: Design a performance benchmark for an API — test scenarios, metrics, and tooling.
- Bench Compare: Compare API performance across versions — regression detection and root cause analysis.
- Bench Recon: Audit existing performance testing — find missing benchmarks, stale baselines, and CI gaps.
Key Rules
- Metrics: p50, p95, p99 latency; requests/second throughput; error rate under load
- Tools: k6 for scripted load tests, wrk for raw throughput, hey for quick HTTP benchmarks
- Baseline: establish baseline on every release; alert on >10% p99 regression
- Realistic payloads: benchmark with production-sized request bodies, not empty payloads
- Warmup: always include a warmup period to fill connection pools and caches
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.
- yesterday First seen · 58 lines · 15 tokens per session scan A e20312123434
bench is an agent published in the GitHub repository tonone-ai/tonone (71 stars, last pushed 16d ago), licensed MIT. It adds 15 tokens to every session and 591 once invoked, about $0.0001 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 2026-09-01.
Other agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
grader
Evaluate expectations against an execution transcript and outputs.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.