benchmark

benchmark is a command for Claude Code from sairam0424/MindForge. It costs 6 tokens per session (289 once invoked), scanned A, original, MIT.

A measurement command for evaluating how well an agent skill performs over time. It compares usage, verification results, detected problems, and session quality.

In plain words
What is it for?
It helps analyze audit and usage logs, compare two skills, and report load frequency, verification pass rates, anti-pattern detection, and context costs.
Why use it?
It helps teams decide whether a skill improves results enough to keep, combine, or replace it.

Command for Claude Code

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.

agentmods
npx agentmods add commands/sairam0424/mindforge/benchmark
Clone the repo
git clone --depth 1 https://github.com/sairam0424/MindForge

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 benchmark

README.md
[![agentmods](https://agentmods.dev/badge/commands/sairam0424/mindforge/benchmark.svg)](https://agentmods.dev/commands/sairam0424/mindforge/benchmark)
Your own site
<a href="https://agentmods.dev/commands/sairam0424/mindforge/benchmark"><img src="https://agentmods.dev/badge/commands/sairam0424/mindforge/benchmark.svg" alt="Measured on agentmods" height="20"></a>
Per session 6 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 289 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin unknown 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 $0.00006 $0.00289
Opus 5 $0.00003 $0.00144
Sonnet 5 $0.00001 $0.00058
Haiku 4.5 $0.00001 $0.00029

Measured today against content hash de66353ef846, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

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 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.

.claude/commands/mindforge/benchmark.md · 38 lines

What it actually says

MindForge — Benchmark Command

Usage: /mindforge:benchmark [--skill skill-name] [--compare skill-a skill-b]

Measure skill effectiveness over time.

Single skill benchmark

For a named skill, analyse AUDIT.jsonl and skill-usage.jsonl:

  • How many times was the skill loaded this month?
  • What is the verify pass rate for tasks where this skill was loaded?
  • Are there anti-patterns less common after this skill is loaded?
  • What is the average session quality score when this skill is active?

Report:

Skill Benchmark: security-review v1.0.0
────────────────────────────────────────
Usage (last 30 days): 47 task loads
Trigger distribution: text match 68%, file-path 22%, file-name 10%
Verify pass rate:     91% (vs. 84% baseline without this skill)
Security findings:    8 HIGH caught (0 CRITICAL missed in tasks using this skill)
Session quality lift: +6.2 points average when loaded

Assessment: HIGH VALUE — clear quality improvement signal

Skill comparison

Compare two skills head-to-head:

  • Load frequency
  • Verify pass rate improvement
  • Anti-pattern detection rate
  • Context budget cost (token estimate)

Helps decide: should you keep both skills, or deprecate the lower-performer?

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. today First seen · 38 lines · 6 tokens per session scan A de66353ef846

Subscribe to this mod's changes

benchmark is a command published in the GitHub repository sairam0424/MindForge (0 stars, last pushed today), licensed MIT. It adds 6 tokens to every session and 289 once invoked, about $0.0000 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.