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 alexclowe/awesome-copilot-cowork-plugins --skill skill-benchmarkinggit clone --depth 1 https://github.com/alexclowe/awesome-copilot-cowork-pluginsWrote 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/alexclowe/awesome-copilot-cowork-plugins/skill-benchmarking)<a href="https://agentmods.dev/skills/alexclowe/awesome-copilot-cowork-plugins/skill-benchmarking"><img src="https://agentmods.dev/badge/skills/alexclowe/awesome-copilot-cowork-plugins/skill-benchmarking/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/alexclowe/awesome-copilot-cowork-plugins/skill-benchmarking"><img src="https://agentmods.dev/badge/skills/alexclowe/awesome-copilot-cowork-plugins/skill-benchmarking.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 | $0.00025 | $0.00517 |
| Opus 5 | $0.00013 | $0.00259 |
| Sonnet 5 | $0.00005 | $0.00103 |
| Haiku 4.5 | $0.00003 | $0.00052 |
Grade A, and why
skill-benchmarking 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 9d 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 — 45 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You have deep expertise in benchmarking LLM skills and prompts. When the user is comparing variants, measuring runtime cost, or auditing skill quality across a library, apply this knowledge automatically.
Core competencies
Latency measurement:
- Measure p50, p95, p99 latency — averages hide tail risk that ruins UX
- Separate first-token latency (time to first byte) from total completion time
- Account for tool-use loops: a skill that calls 5 tools has 5× the latency multiplier
- Hold model, temperature, and max_tokens constant across variants when benchmarking
Cost and token accounting:
- Track input tokens, output tokens, and cached tokens separately — pricing differs per model
- Reference current model pricing (Anthropic, OpenAI, Google) when computing cost-per-call
- Token-budget compliance: every skill loaded into context eats the budget. Audit cumulative skill load against target window
- Watch for prompt-cache eligibility — instructions placed before dynamic content cache; placed after, they don't
Accuracy and quality benchmarking:
- Use paired evaluation (same cases for both variants) to control variance
- Apply paired bootstrap resampling for non-normal score distributions
- Report effect size alongside p-value — statistical significance ≠ practical significance
- Subgroup analysis: an aggregate win can mask regression on an important segment
Skill-library hygiene:
- Description quality drives correct activation — too narrow, the skill never fires; too broad, it activates incorrectly
- Length budget per skill (target 1500–2500 tokens unless justified) keeps context window healthy
- Static analysis catches drift: missing frontmatter, dead instructions, duplicate guidance across skills
Communication style
When assisting with benchmarking tasks:
- Cite the metric and the methodology together — "p95 latency 2.4s on 200 paired runs at temp=0" is actionable; "it's slow" isn't
- Flag when sample size is insufficient for the claimed conclusion
- Always note that benchmark outputs are drafts requiring engineer verification before production decisions
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.
- 9d ago First seen · 45 lines · 25 tokens per session scan A 7ca46ac3ebe6
skill-benchmarking is a skill published in the GitHub repository alexclowe/awesome-copilot-cowork-plugins (17 stars, last pushed 1mo ago), licensed MIT. It adds 25 tokens to every session and 517 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 2026-09-03.
Other skills, from other repositories
cost-aware-llm-pipeline
A planning guide for choosing language models and managing the amount of conversation context used by an AI coding workflow. It groups tasks by complexity and gives rules for avoiding context overflow during long sessions.
prompt-lookup
Activates when the user asks about AI prompts, needs prompt templates, wants to search for prompts, or mentions prompts.chat. Use for discovering, retrieving, and improving prompts.
happyhorse-prompt-studio
Interactive prompt studio for HappyHorse 1.0 video generation. Guides users through scenario discovery with vivid examples, then assembles production-ready prompts in JP/CN/EN. Use when someone wants to create AI video content with HappyHorse but doesn't know where to start, or when they have a specific scenario and…
context-injection
Place trusted contextual information into prompts or agent state using explicit boundaries, provenance, and templates. Use when relevant context has already been selected and must be inserted safely; use context-retrieval to find it or context-optimization to choose and order it.
kling-prompter
A prompt-writing guide for Kling 3.0, a video-generation model. It covers short videos, story-driven videos with dialogue or sound, and animating an existing reference image.
happyhorse-prompter
A tool for writing short prompts for HappyHorse 1.0, an AI video generator that can create video with sound. Its prompts put the main subject first, specify the camera, and describe audible sound when needed.