tfactory-watch
Pass
Audited by Gen Agent Trust Hub on Jun 20, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill ingests and processes untrusted data from the TFactory autonomous pipeline output, creating a surface for indirect prompt injection.
- Ingestion points: The agent reads
triage_report.md,logs/*.log, andreplan_request.jsonfrom the local workspace. - Boundary markers: There are no delimiters or specific instructions (e.g., 'ignore embedded commands') provided to the agent to separate the report content from its primary instructions.
- Capability inventory: The agent's access to the
Bashtool and variousmcp__tfactory__*tools increases the potential impact if it follows instructions embedded in the reports. - Sanitization: No content validation or sanitization is performed on the files read from the workspace before the agent analyzes them.
Audit Metadata