Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
Eidolon Search
v0.0.1AI Agent memory search using SQLite FTS5. 90%+ token reduction (10x+) compared to reading full files. Use when the agent needs to search through markdown mem...
⭐ 0· 290·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description (FTS5 memory search) align with included scripts (build-index.py, search.py, benchmark scripts). Files implement indexing, searching, and local benchmarks; nothing requests unrelated cloud credentials or services.
Instruction Scope
SKILL.md instructs the agent to run local Python scripts that read .md files from a specified memory directory and write a local SQLite DB. This is within the stated purpose. Note: benchmark-cache.py attempts to drop the OS page cache by invoking sudo sync and writing to /proc/sys/vm/drop_caches (Linux-only); that requires elevated privileges if actually run and is only relevant to optional benchmarking.
Install Mechanism
No install spec; the skill is instruction + local Python scripts. No downloads or third-party package installs are invoked by the skill itself.
Credentials
The skill requests no environment variables or credentials; search.py optionally reads DB_PATH from the environment which is reasonable and proportional. There are no requests for unrelated secrets or config paths.
Persistence & Privilege
always is false and the skill does not request persistent or elevated platform privileges. It creates/uses a local DB file (memory.db) within the user's workspace as expected; it does not modify other skills or system-wide agent settings.
Assessment
This appears to be a straightforward local FTS5-based search tool. Before installing/running: (1) ensure you only index directories you intend to (the index will read and store the contents of all matched .md files into memory.db); (2) the benchmark script tries to drop the OS page cache using sudo—only run it if you understand and allow that; (3) confirm your Python/SQLite build includes FTS5 support; (4) there is no network exfiltration or credential use in the code, but treat the created DB as sensitive if it contains private notes and store it appropriately. If you want the agent to run this autonomously, be aware the agent could call the included scripts and thus read any files in the provided memory directory.Like a lobster shell, security has layers — review code before you run it.
latestvk97bsya11y4ra9xqc481m3qs7s82a7kz
License
MIT-0
Free to use, modify, and redistribute. No attribution required.