auth-security-expert

Pass

Audited by Gen Agent Trust Hub on Mar 3, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The SKILL.md file contains a mandatory 'Memory Protocol' instruction that directs the agent to execute a bash command (cat .claude/context/memory/learnings.md) to retrieve state information at the beginning of a session.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests data from .claude/context/memory/learnings.md and treats it as instruction context without validation or boundary delimiters.\n
  • Ingestion points: .claude/context/memory/learnings.md (read via the Bash tool).\n
  • Boundary markers: Absent; the file content is processed as direct instructions or context.\n
  • Capability inventory: The skill utilizes the Bash, Write, Edit, and Read tools, which could be leveraged by injected instructions to perform unauthorized file or system modifications.\n
  • Sanitization: Absent; the agent is not instructed to validate or escape the content of the memory file before processing it.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 3, 2026, 02:04 PM