status

Pass

Audited by Gen Agent Trust Hub on Feb 28, 2026

Risk Level: SAFENO_CODEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill instructs the agent to read content from various markdown files (e.g., context-builder.md, jtbd-capture.md) to find status markers. This creates a surface for indirect prompt injection if those files contain malicious instructions meant to hijack the agent's behavior during the status report.
  • Ingestion points: Reads the content of 10+ specific markdown files and scans the context/ directory.
  • Boundary markers: None. The agent is not provided with delimiters or instructions to treat the file content as untrusted data.
  • Capability inventory: None. The skill does not contain any scripts, subprocess calls, or network operations.
  • Sanitization: None. The skill relies on natural language parsing without explicit validation of the file contents.
  • [NO_CODE]: This skill contains no executable scripts (Python, JavaScript, Bash) or binaries. It is composed entirely of markdown instructions for the AI agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 28, 2026, 03:25 AM