Claude Code Guide
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- Prompt Injection (LOW): The skill suggests using the phrase 'forget previous instructions' as a debugging technique. While provided as user advice, this matches the 'ignore-previous-instructions' pattern used to override agent constraints.- Indirect Prompt Injection (LOW): The skill defines a workflow where the agent reads and obeys instructions from a local
CLAUDE.mdfile. This is a data ingestion surface that could be exploited if the file content is attacker-controlled. (Ingestion points:CLAUDE.md; Boundary markers: absent; Capability inventory: none; Sanitization: absent).- No Code (SAFE): The skill consists entirely of documentation and markdown templates with no executable scripts, reducing the risk of direct command execution or malware.
Audit Metadata