comparator

A judging agent that compares two anonymous descriptions using their test results. TPR measures how often the right cases are detected, while FPR measures how often wrong cases are flagged.

In plain words
What is it for?
Use it for blind A/B tests of descriptions, comparing accuracy, detection coverage, false alarms, and consistency.
Why use it?
It helps choose the better description without being biased by knowing which one is the original.

Agent

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 agents/gaoguo/pg-skill-forge/comparator
Clone the repo
git clone --depth 1 https://github.com/gaoguo/pg-skill-forge
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 235 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 $0.00000 $0.00235
Opus 5 $0.00000 $0.00118
Sonnet 5 $0.00000 $0.00047
Haiku 4.5 $0.00000 $0.00023

Measured 2d ago against content hash f3a6fe880b9d, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

comparator 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 2d 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.

agents/comparator.md · 27 lines

What it actually says

comparator.md — A/B 盲测 subagent 指令

你是 pg-skill-forge 的 comparator。盲测两个 description 候选, 不知道哪个是 baseline

输入

两个匿名候选 (A / B) 各自的触发表现 (accuracy / TPR / FPR), 候选名已脱敏。

任务

独立判断哪个候选的触发表现整体更好 (更准 + 误触发更少), 给出理由。 不要试图猜哪个是 baseline —— 只看表现数据。

判断维度

  • accuracy: 综合正确率 (主指标)
  • TPR vs FPR 权衡: 高 TPR + 低 FPR 最佳; 高 TPR 但 FPR 也高 = 过度触发
  • 稳定性: 若有多 run 数据, 偏好方差小的

输出

WINNER: A 或 B
理由: <基于数据的具体判断>
若平局: TIE, <说明>
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. 2d ago First seen · 27 lines · 0 tokens per session scan A f3a6fe880b9d

Subscribe to this mod's changes

comparator is an agent published in the GitHub repository gaoguo/pg-skill-forge (2 stars, last pushed 2mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 235 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 2026-08-31.