recoup-internal-consulting-stakeholder-update
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill consists entirely of markdown instructions defining a specific tone and structure for stakeholder communications. It does not include any scripts, commands, or automated triggers.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to summarize and process information from the agent's local workspace.
- Ingestion points: Reads project-level data including
AGENTS.mdand entity folders as described in the workspace instructions inSKILL.md. - Boundary markers: The instructions include a 'Gut check before sending' review section and reference an external gating process ('gate before sending') to ensure content accuracy.
- Capability inventory: The skill's primary capability is text generation for drafts; it does not perform file writes, network requests, or shell executions.
- Sanitization: The skill relies on human-in-the-loop review ('read it as them') and structural gating rather than automated technical sanitization.
Audit Metadata