skill-creation
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the analysis of past session logs located in
~/.factory/sessions/*.jsonlto extract patterns and codify them into new skills. This ingestion of historical interaction data creates a surface for indirect prompt injection if those logs contain malicious instructions designed to be 'learned' and persistent in future skills. - Ingestion points: Historical session log files (
~/.factory/sessions/*.jsonl). - Boundary markers: Absent; the skill relies on the agent's natural language ability to generalize and extract logic without explicit delimiters for untrusted content.
- Capability inventory: The skill utilizes file system write capabilities to create or update markdown files in
.factory/skills/and~/.factory/skills/. - Sanitization: No explicit sanitization or validation steps are provided for the content extracted from sessions before it is written into new skill instructions.
- [COMMAND_EXECUTION]: The instructions direct the agent to execute shell commands such as
rg(ripgrep) to search through session files. This enables the agent to access and process the contents of all previous session data stored on the local filesystem.
Audit Metadata