brief

Pass

Audited by Gen Agent Trust Hub on Sep 20, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill performs expected file system operations, including reading company context from ~/.claude/company-context.md and saving generated briefs to ~/.claude/briefs/. These actions are restricted to the agent's local environment and are necessary for the skill's primary function.
  • [INDIRECT_PROMPT_INJECTION]: The skill acts as a data ingestion surface by processing user-provided topics or office-hours intake. While this data is interpolated into the strategy brief, the risk is negligible as the output is a static Markdown file and no dangerous capabilities (such as command execution or network exfiltration) are triggered by the ingested content.
  • Ingestion points: User input through the <topic> or <office-hours-output> parameters in SKILL.md.
  • Boundary markers: No specific delimiters or warnings for embedded instructions are defined.
  • Capability inventory: Writing Markdown files to the local file system (~/.claude/briefs/).
  • Sanitization: No explicit sanitization or validation of the intake data is mentioned.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 20, 2026, 09:19 AM
Security Audit — agent-trust-hub — brief