benchmark

A command that compares a skill's historical evaluation scores with defined ideal benchmarks. A benchmark is a target used to judge whether performance is on track.

In plain words
What is it for?
Use it to review average scores across evaluation dimensions and identify areas where a skill needs improvement.
Why use it?
It shows which quality areas are below, at, or above the expected target instead of leaving historical scores without context.

Command

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/sattyamjjain/proofloop/benchmark
Clone the repo
git clone --depth 1 https://github.com/sattyamjjain/proofloop
Per session 8 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 661 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00008 $0.00661
Opus 5 $0.00004 $0.00331
Sonnet 5 $0.00002 $0.00132
Haiku 4.5 $0.00001 $0.00066

Measured yesterday against content hash f9d2780459ce, 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 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.

commands/benchmark.md · 65 lines

How it starts

The opening of the file, as written. The whole thing — 65 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/benchmark — Compare Against Ideal Benchmarks

Compare a skill's historical performance against defined benchmark standards.

Arguments

  • skill-name (required): The skill to benchmark.

What to Do

  1. Read benchmark standards from skills/judge/references/benchmark-standards.md
  2. Read historical scores for the specified skill from skills/judge/scores/
  3. Compute average per dimension across all historical evaluations
  4. Compare against benchmark:
┌───────────────────────────────────────────────────────────────┐
│  PROOFLOOP BENCHMARK — {skill-name}                          │
├────────────────┬──────────┬───────────┬───────────────────────┤
│ Dimension      │ Your Avg │ Benchmark │ Delta                 │
├────────────────┼──────────┼───────────┼───────────────────────┤
│ Correctness    │ 8.2      │ 8.5       │ -0.3 (Below)          │
│ Completeness   │ 7.5      │ 8.0       │ -0.5 (Below)          │
│ Adherence      │ 9.0      │ 8.0       │ +1.0 (Above)          │
│ Actionability  │ 8.0      │ 8.0       │  0.0 (On target)      │
│ Efficiency     │ 7.0      │ 7.5       │ -0.5 (Below)          │
│ Safety         │ 9.5      │ 9.0       │ +0.5 (Above)          │
│ Consistency    │ 6.5      │ 7.0       │ -0.5 (Below)          │
├────────────────┼──────────┼───────────┼───────────────────────┤
│ COMPOSITE      │ 8.05     │ 8.14      │ -0.09                 │
└────────────────┴──────────┴───────────┴───────────────────────┘

Strengths: Adherence (+1.0), Safety (+0.5)
Weaknesses: Completeness (-0.5), Efficiency (-0.5), Consistency (-0.5)

Recommendations:
1. Focus on covering all requirements (Completeness gap)
2. Reduce unnecessary tool calls (Efficiency gap)
3. Build more consistent quality (Consistency gap)

If no scores exist for the skill, inform the user to run /judge first.

Regression benchmark (CI gate)

The /benchmark command above compares a skill's own history to reference standards. For heuristic-regression testing, use scripts/benchmark_pack.py instead:

Read the full file on GitHub · 65 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. yesterday First seen · 65 lines · 8 tokens per session scan A f9d2780459ce

Subscribe to this mod's changes

benchmark is a command published in the GitHub repository sattyamjjain/proofloop (5 stars, last pushed 2mo ago), licensed MIT. It adds 8 tokens to every session and 661 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 2026-08-31.