Skills
skills/first-fluke/oh-my-ag/oma-recap/Gen Agent Trust Hub

oma-recap

Pass

Audited by Gen Agent Trust Hub on May 3, 2026

Risk Level: SAFEDATA_EXFILTRATIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [DATA_EXFILTRATION]: The skill accesses conversation history files located at ~/.claude/history.jsonl. This data contains previous interactions and is accessed to generate work summaries. No network transmission of this data was detected.
  • [COMMAND_EXECUTION]: The skill uses the oma CLI tool and jq for processing history data.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted conversation history data. Malicious instructions embedded in past conversations could potentially influence the agent's summary generation process.
  • Ingestion points: ~/.claude/history.jsonl and output from the oma recap command.
  • Boundary markers: None identified in the prompt interpolation process.
  • Capability inventory: Command execution via oma and jq, and file writing to the .agents/results/recap/ directory.
  • Sanitization: No explicit sanitization or filtering of historical prompt content is described.
Audit Metadata
Risk Level
SAFE
Analyzed
May 3, 2026, 09:10 AM