dev3-prd

Pass

Audited by Gen Agent Trust Hub on Aug 14, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data from files that may contain untrusted content, such as feature requests or codebase exploration results.
  • Ingestion points: Reads feature context and design details from .llm/$ARGUMENTS/tmp_context.md, .llm/$ARGUMENTS/tmp_exploration.md, and .llm/$ARGUMENTS/design.md.
  • Boundary markers: None. The skill does not define clear delimiters or provide instructions to the agent to ignore embedded commands within the ingested data.
  • Capability inventory: The skill can write files to the local .llm directory and trigger the next step in the automated sequence via the /dev4-implement command.
  • Sanitization: No sanitization or content validation is performed on the data read from the input files before it is used to generate documentation or drive subsequent tool calls.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 14, 2026, 12:37 AM
Security Audit — agent-trust-hub — dev3-prd