remind
Pass
Audited by Gen Agent Trust Hub on Mar 10, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: A detailed review of
scripts/search.py,SKILL.md, and metadata found no evidence of malicious code, obfuscation, or unauthorized connectivity. - [DATA_EXFILTRATION]: Analyzed file access patterns in
scripts/search.py. The script reads conversation logs from~/.claude/projects/and~/.scribe/scribe.db. This access is necessary for the skill's intended purpose of recalling past sessions, and no network capabilities are present to exfiltrate this data. - [PROMPT_INJECTION]: Evaluated the surface for indirect prompt injection as the skill processes past conversation content. Ingestion points: Reads conversation text from local JSONL logs and SQLite database files. Boundary markers: None explicitly present in prompt interpolation. Capability inventory: Checked
scripts/search.py; no subprocess calls,exec/eval, file-write operations, or network requests were found. Sanitization: Thescribe_searchfunction utilizes regex to strip HTML/XML tags and ANSI escape sequences from the retrieved content.
Audit Metadata