skill-suggester
Pass
Audited by Gen Agent Trust Hub on Jun 23, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection by processing untrusted data from local conversation logs.
- Ingestion points: Accesses local history files at
~/.local/state/opencode/prompt-history*.jsonlas defined in SKILL.md. - Boundary markers: No delimiters or specific 'ignore instructions' markers are defined to isolate the ingested log content from the agent's primary instructions.
- Capability inventory: Reads local files and incorporates their content into the active session for pattern analysis and summarization.
- Sanitization: The skill relies on natural language instructions for the agent to 'summarize patterns' and avoid private excerpts, but lacks technical sanitization or validation of the input data.
Audit Metadata