← 返回 Skills 市场

AEO Prompt Frequency Analyzer

Name: AEO Prompt Frequency Analyzer
Author: psyduckler

作者 psyduckler · GitHub ↗ · v1.0.0

cross-platform ⚠ suspicious

756

总下载

当前安装

版本数

在 OpenClaw 中安装

/install aeo-prompt-frequency-analyzer

功能描述

Analyze what search queries Gemini uses when answering a prompt, by running it multiple times with Google Search grounding and reporting frequency distributi...

使用说明 (SKILL.md)

Prompt Frequency Analyzer

Run a prompt N times against Gemini with Google Search grounding enabled. Collect and report the frequency of search queries Gemini generates across all runs.

Usage

GEMINI_API_KEY=$(security find-generic-password -s "nano-banana-pro" -w) \
  python3 scripts/analyze.py "your prompt here" [--runs 10] [--model gemini-2.5-pro] [--concurrency 5] [--output text|json]

Run from the skill directory. Resolve scripts/analyze.py relative to this SKILL.md.

Options

--runs N — Number of times to run the prompt (default: 10)
--model NAME — Gemini model to use (default: gemini-2.5-pro)
--concurrency N — Max parallel API calls (default: 5; keep ≤5 to avoid rate limits)
--output text|json — Output format (default: text)

Output

Reports for each unique search query:

Frequency percentage (how many runs used that query)
Raw count
Top web sources referenced

Notes

Gemini API key must be in GEMINI_API_KEY env var (stored in macOS Keychain under nano-banana-pro)
Each run is independent — Gemini may use different search queries each time
Retries failed requests up to 3 times with exponential backoff
Use --output json for programmatic consumption

安全使用建议

This skill's code matches its description: it repeatedly calls the Gemini API with Google Search grounding and aggregates queries. However, the registry metadata does not declare that GEMINI_API_KEY is required, while SKILL.md and the script both expect it — SKILL.md even suggests retrieving it from macOS Keychain using a specific key name ('nano-banana-pro'). Before installing or running: 1) Verify the Gemini API key source and name (you may prefer to set GEMINI_API_KEY explicitly rather than using the sample keychain command). 2) Be aware the script will make multiple outbound requests to Google's API (costs and rate limits possible). 3) If you are not on macOS, the suggested 'security' command won't apply; adjust instructions accordingly. 4) Because the package source/homepage is unknown, consider auditing the included script (scripts/analyze.py) yourself or running it in an isolated environment. The mismatches between registry metadata and SKILL.md lower trust but do not by themselves indicate malicious behavior.

功能分析

Type: OpenClaw Skill Name: aeo-prompt-frequency-analyzer Version: 1.0.0 The skill bundle is benign. Its purpose is to analyze Gemini's search query patterns for a given prompt by repeatedly calling the Gemini API. The `SKILL.md` explicitly instructs the agent to retrieve the `GEMINI_API_KEY` from the macOS Keychain, which is then used by `scripts/analyze.py` to interact with the legitimate `generativelanguage.googleapis.com` endpoint. All actions, including credential handling and network calls, are transparent, align with the stated purpose, and show no evidence of data exfiltration, malicious execution, persistence mechanisms, or prompt injection attempts against the OpenClaw agent itself.

能力评估

ℹ Purpose & Capability

The code and SKILL.md implement exactly what the skill claims: repeated calls to Gemini with Google Search grounding and frequency reporting. That behavior is coherent with the name/description. However, the registry metadata claims no required environment variables even though the script and SKILL.md require a Gemini API key, which is an inconsistency.

⚠ Instruction Scope

SKILL.md instructs users to fetch GEMINI_API_KEY from the macOS Keychain using the 'security' command and a specific key name ('nano-banana-pro'). The script itself only reads GEMINI_API_KEY from the environment. The SKILL.md's macOS-specific retrieval step and hard-coded key name are undocumented in the registry metadata and expand the practical scope (requires a keychain entry and the 'security' binary) without declaration.

✓ Install Mechanism

This is an instruction-only skill with a bundled Python script and no install spec. Nothing is downloaded or executed beyond the included script and standard Python/urllib usage, which is low-risk from an install perspective.

⚠ Credentials

The script legitimately needs a single GEMINI_API_KEY to call the Google generativelanguage API. But the registry metadata lists no required env vars while SKILL.md and the script require GEMINI_API_KEY (and even recommend pulling it from a specific keychain entry). The request for an API key is proportionate to the task, but the missing declaration and the hard-coded keychain name are red flags (metadata mismatch, OS-specific guidance).

✓ Persistence & Privilege

The skill does not request permanent/always-loaded presence, does not modify other skills or system-wide configs, and does not store credentials itself. It simply makes outbound API calls when run.

如何使用

确保已安装 OpenClaw（本地或 Docker 部署）
在对话框中输入安装命令：/install aeo-prompt-frequency-analyzer
安装完成后，直接呼叫该 Skill 的名称或使用 /aeo-prompt-frequency-analyzer 触发
根据 Skill 的参数说明提供必要输入，即可获得结构化输出

版本历史

v1.0.0

Initial publish

元数据

Slug aeo-prompt-frequency-analyzer

版本 1.0.0

许可证 —

累计安装 1

当前安装数 1

历史版本数 1

常见问题