technical-analyst
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data which can potentially influence the agent's behavior.
- Ingestion points: The skill ingests user-provided chart images as specified in SKILL.md and loads a JSON report via the --detector-json CLI argument in scripts/check_weekly_price_action.py.
- Boundary markers: The skill uses a structured technical analysis framework and report templates to guide output, but lacks explicit ignore-instructions boundary markers for the external JSON data.
- Capability inventory: The skill performs network operations via requests.get to a financial API and writes analysis reports to the local file system.
- Sanitization: The Python logic performs type and key existence checks on the JSON input before using it to determine market direction.
- [COMMAND_EXECUTION]: The skill's test suite utilizes shell command execution.
- Evidence: The file scripts/tests/test_check_weekly_price_action.py uses subprocess.run() to execute the skill's own CLI for functional testing.
Audit Metadata