skill-benchmark

skill-benchmark is a command for Claude Code from HoangNguyen0403/agent-skills-standard. It costs 0 tokens per session (669 once invoked), scanned A, original, MIT.

A benchmarking workflow that measures how much a set of agent skills improves implementation quality against known legacy problems. It creates a scorecard and identifies where the skills help or fall short.

In plain words
What is it for?
Use it to compare implementations with and without active skills and prioritize improvements to those skills.
Why use it?
It provides evidence about whether the skills are useful instead of relying on opinion. It can reveal recurring code-quality problems that the guidance does not prevent.

Command for Claude Code

Written for Claude Code: $ARGUMENTS substitution. Also seen: mentions AGENTS.md.

Good fit Use it to compare implementations with and without active skills and prioritize improvements to those skills.

Compare 6 commands from other repositories ↓
Install with agentmods
npx agentmods add commands/hoangnguyen0403/agent-skills-standard/skill-benchmark
Install

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.

Clone the repo
git clone --depth 1 https://github.com/HoangNguyen0403/agent-skills-standard

Made for: Claude Code.

Wrote 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.

agentmods badge for skill-benchmark

README.md
[![agentmods](https://agentmods.dev/badge/commands/hoangnguyen0403/agent-skills-standard/skill-benchmark/github.svg)](https://agentmods.dev/commands/hoangnguyen0403/agent-skills-standard/skill-benchmark)
Your own site
<a href="https://agentmods.dev/commands/hoangnguyen0403/agent-skills-standard/skill-benchmark"><img src="https://agentmods.dev/badge/commands/hoangnguyen0403/agent-skills-standard/skill-benchmark/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.

agentmods 80×15 button for skill-benchmark

Your own site · 80×15
<a href="https://agentmods.dev/commands/hoangnguyen0403/agent-skills-standard/skill-benchmark"><img src="https://agentmods.dev/badge/commands/hoangnguyen0403/agent-skills-standard/skill-benchmark.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 669 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00000 $0.00669
Opus 5 $0.00000 $0.00334
Sonnet 5 $0.00000 $0.00134
Haiku 4.5 $0.00000 $0.00067

Measured 9d ago against content hash 1e15a3f3abe0, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

skill-benchmark 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.

.claude/commands/skill-benchmark.md · 84 lines

How it starts

The opening of the file, as written. The whole thing — 84 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Skill Benchmark

Benchmark AI skill effectiveness by measuring implementation quality against legacy constraints.

Input: $ARGUMENTS

Optional args: slug=, ticket=<id/url>, mode=interactive|autonomous|channel, channel=, auto_continue=true|false, profile=business|hybrid|technical.

Instructions

Execute the following steps for $ARGUMENTS.

📊 Skill Benchmark Orchestrator

Goal: Quantify how much active skills improve implementation quality. Deliver a prioritized compliance delta and skill applicability report.


Step 1 — Project Context & Active Skills

Identify the tech stack and all active skills in AGENTS.md.

# 1. Total source files and lines changed
find src -name "*.ts" -o -name "*.tsx" | xargs wc -l 2>/dev/null | sort -rn | head -20
# 2. Check active skill registry
cat AGENTS.md | head -80

Step 2 — Auto-Select a Legacy Trap

Pick the file automatically. Rank candidates by the severity of anti-patterns:

  • 🔴 P0: Hardcoded secrets; Logic inside UI components.
  • 🟠 P1: Wrong Router pattern; Global state for local concerns; Missing design tokens.
  • 🟡 P2: Raw user-facing strings (i18n).

Step 3 — Build Eval-Driven Scorecard

Source your scorecard from evals/evals.json, not from hardcoded patterns. Follow the Scorecard Rubric in <SKILLS>/common/common-skill-creator/references/benchmark.md when synced:

  1. Read <SKILLS>/<category>/<skill>/evals/evals.json.
  2. Generate columns for Failure Pattern and Success Pattern.
  3. Refactor the file, citing the exact skill rule for each change.
  4. For guardrail skills, read pressure_scenarios, rationalizations, red_flags, and behavior_assertions.

Step 4 — Benchmark Report & Compliance Delta

Output the scorecard and compliant score using the templates in <SKILLS>/common/common-skill-creator/references/benchmark.md when synced.

  • Compliance Score Before vs After.
  • Δ Delta: +Z% 🚀.
  • Eval Alignment: How well does the skill teach what the eval tests?
  • Behavior Coverage: pressure scenarios, rationalizations, red flags, behavior assertions.

Read the full file on GitHub · 84 lines

Changes

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.

  1. 9d ago First seen · 84 lines · 0 tokens per session scan A 1e15a3f3abe0

Subscribe to this mod's changes

skill-benchmark is a command published in the GitHub repository HoangNguyen0403/agent-skills-standard (565 stars, last pushed 3d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 669 tokens. 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.