analysis-history

Pass

Audited by Gen Agent Trust Hub on Mar 20, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the Bash tool to execute python3 scripts to process and compare historical artifacts such as state.json, report.md, and metadata.json found in the runtime tree.
  • [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface by processing data from previous analysis runs that may contain content from untrusted external sources. 1. Ingestion points: Reads artifact files (state.json, report.md) from the runtime 'runs/' tree. 2. Boundary markers: Absent; there are no specific instructions or delimiters defined to prevent the agent from obeying instructions embedded within these historical files. 3. Capability inventory: The skill has access to file system tools (Read, Glob) and the ability to execute code via Bash/Python. 4. Sanitization: Absent; the skill does not explicitly define methods for sanitizing or escaping the content of the historical artifacts before they are presented to the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 20, 2026, 06:01 AM