What is prompt-evaluator?

A prompt-testing agent evaluates AI prompts against defined success criteria using promptfoo, a tool for running repeatable prompt tests. It records scores and can stop prompts from being published when they miss the required threshold.

It is for testing prompts, generating test files, scoring results across specified AI providers, and saving evaluation details in prompt metadata.

It replaces subjective judgments with repeatable test results and keeps untested or low-scoring prompts out of the library.

prompt-evaluator — Agent for Claude Code by frankxai

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/frankxai/agentic-creator-os

Made for: Claude Code.

Or install agentic-creator-os, the plugin that ships this one along with the rest of its 133 commands, 68 agents.

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.

README.md

[![agentmods](https://agentmods.dev/badge/agents/frankxai/agentic-creator-os/prompt-evaluator/github.svg)](https://agentmods.dev/agents/frankxai/agentic-creator-os/prompt-evaluator)

Your own site

<a href="https://agentmods.dev/agents/frankxai/agentic-creator-os/prompt-evaluator"><img src="https://agentmods.dev/badge/agents/frankxai/agentic-creator-os/prompt-evaluator/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.

Your own site · 80×15

<a href="https://agentmods.dev/agents/frankxai/agentic-creator-os/prompt-evaluator"><img src="https://agentmods.dev/badge/agents/frankxai/agentic-creator-os/prompt-evaluator.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 114 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 1,233 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.

Model	Per session	Once invoked
Fable 5.1	$0.00114	$0.01233
Opus 5	$0.00057	$0.00616
Sonnet 5	$0.00023	$0.00247
Haiku 4.5	$0.00011	$0.00123

Measured 9d ago against content hash f9a3f992f52f, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

prompt-evaluator 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/agents/prompt-evaluator.md · 120 lines

How it starts

The opening of the file, as written. The whole thing — 120 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Prompt Evaluator

Mission

Evaluate a prompt against declared success criteria. Score it. Write the score back to the pattern's frontmatter. Block publish if the score is below threshold.

No assertion of "this is better" without numbers.

When to invoke

@prompt-conductor dispatches flow-evaluate.
Inside flow-design, flow-optimize, flow-curate as the score gate.
"evaluate this prompt", "test this system prompt", "score this prompt", "run evals on X".

Hard rules

Wrap promptfoo, don't reinvent it. npx promptfoo eval --config <path>.
Every pattern gets a colocated evals/promptfoo.yaml. No untested prompts in prompt-library.
Score threshold for publish: ≥3.5/5 weighted. Below threshold, return to optimizer.
Brand voice check is an eval assertion, not a separate step. not-contains: ["delve", "dive into", ...].
Run on at least 2 lab providers when the pattern's lane is cross-lab. Otherwise the declared lane.
Persist the score to pattern.md frontmatter eval.score, eval.last_run, eval.test_count.
No vibes scores. Every score is npx promptfoo eval output.

Promptfoo file template

description: Evals for <pattern_id>
providers:
  - anthropic:messages:claude-opus-4-7
  - openai:gpt-5                          # only if cross-lab
  - google:gemini-3-pro                   # only if cross-lab
prompts:
  - file://../pattern.md
defaultTest:
  assert:
    # brand voice gate
    - type: not-contains
      value: ["delve", "dive into", "certainly", "absolutely", "it's worth noting"]
    # length sanity
    - type: javascript
      value: "output.length < 4000"
tests:
  - vars:
      input: <sample input 1>
    assert:
      - type: contains-all
        value: <required-tokens>
      - type: llm-rubric
        rubric: |
          Score output 1-5 against these criteria:
          - Accurate to <success_criterion>
          - Follows declared output format
          - No hallucination
        threshold: 0.8
  - vars:
      input: <sample input 2>
    assert: ...

Read the full file on GitHub · 120 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.

9d ago First seen · 120 lines · 114 tokens per session scan A f9a3f992f52f

Subscribe to this mod's changes

prompt-evaluator is an agent published in the GitHub repository frankxai/agentic-creator-os (10 stars, last pushed yesterday), licensed Apache-2.0. It adds 114 tokens to every session and 1,233 once invoked, about $0.0006 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.

Claude Code Data and AI Multi-Agent Prompting Testing agentic-skills ai-agents autonomous-coding claude-code-plugins

ai-evaluator

Designs and runs AI product evaluation frameworks: error analysis, eval suite design, LLM-as-judge pipelines, human eval protocols, regression testing plans, and improvement flywheels. Use this agent when the user is building an AI-powered feature and needs to define how to measure quality, catch regressions, or…

Productfculty-aipm/PM-Copilot-by-Product-Faculty · 282 tokens

prompt

Designs versioned system prompts, few-shot libraries, and chain-of-thought patterns with A/B testing and regression coverage — treats prompts as production code. Use when engineering a production LLM feature, auditing a prompt library for drift, or building prompt versioning infrastructure. Trigger with "design this…

jeremylongshore/tons-of-skills-marketplace · 70 tokens

prompt-engineer-pm

Owns the AI product's PROMPT discipline — versioning, registry, prompt-as-code, prompt review, prompt-vs-fine-tune decisions. The PM-side architect for everything the product sends to a model. NOT to be confused with query-refiner-pm (which refines USER queries TO great-pm).

VandanaAjayDubey111/great-pm · 71 tokens

prompt-evaluator