opencode-review-past-sessions

Pass

Audited by Gen Agent Trust Hub on Sep 18, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the pnpm -C packages/reconstituter opencode-sessions command to interact with session logs. This is a legitimate use of a local project tool to recover context from prior sessions.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests data from past sessions which might contain instructions or untrusted input from previous users. * Ingestion points: Historical messages and session excerpts retrieved via the opencode-sessions command as described in SKILL.md. * Boundary markers: The skill does not explicitly define markers to separate historical data from current instructions. * Capability inventory: The agent is limited to searching, identifying relevant session IDs, and summarizing content. * Sanitization: There is no evidence of sanitization for the retrieved session content.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 18, 2026, 10:49 AM
Security Audit — agent-trust-hub — opencode-review-past-sessions