语料检索工具，与 corpus-builder 配合使用。支持语义搜索、元数据过滤（场景/情绪/节奏/质量）。Use when: 需要搜索语料库中的小说片段、按场景类型过滤、查找特定情绪/节奏的描写、检索高质量写作素材。

This skill appears to do what it says: local semantic search over a ChromaDB corpus produced by corpus-builder. Before installing or running: 1) ensure the configured persist_directory points to the corpus you expect (inspect configs/default_config.yml); 2) be aware model loading (sentence-transformers) may download large weights from the internet — run in an environment with sufficient disk space and network policy you control; 3) verify the corpus directory contains only data you're willing to let the skill read (it will access files under the corpus-builder path); 4) optionally run the script in a sandbox or inspect the full script if you want to confirm behavior. The minor issues: requirements.txt includes diskcache although the code currently uses in-memory caching — harmless but worth noting.

Type: OpenClaw Skill Name: corpus-search Version: 1.0.1 The corpus-search skill bundle is a legitimate utility for performing semantic searches and metadata filtering on text corpora. The primary script, scripts/search_corpus.py, uses standard libraries like ChromaDB and SentenceTransformers to query a vector database, with explicit privacy measures such as disabling anonymized telemetry. No indicators of malicious intent, data exfiltration, or prompt injection were found; the code aligns perfectly with its stated purpose of retrieving writing materials.