cline-session-history
Pass
Audited by Gen Agent Trust Hub on Jul 17, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local Python script (scripts/search_sessions.py) to perform search operations. This is the intended method for interacting with the local session store.
- [DATA_EXFILTRATION]: The skill reads sensitive user data, including conversation transcripts and prompts, from the user's local session history directory. This exposure is required for the skill's primary function of browsing history.
- [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface because it retrieves and displays content from past sessions which could contain malicious payloads.
- Ingestion points: The scripts/search_sessions.py script reads data from metadata JSON files, transcript JSON files, and a SQLite database (sessions.db) in the user's session directory.
- Boundary markers: The skill lacks explicit boundary markers to isolate retrieved history from current instructions during agent processing.
- Capability inventory: The skill can read local files, query a SQLite database, and copy files to other local directories (file: scripts/search_sessions.py).
- Sanitization: The script performs basic sanitization by using regex to strip internal tags from transcripts (file: scripts/search_sessions.py, function: print_transcript).
Audit Metadata