eagle-architect
Pass
Audited by Gen Agent Trust Hub on Mar 13, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides a structured methodology for designing system architecture, encompassing territory mapping, domain boundary definition, and architectural documentation. It encourages best practices like using Architecture Decision Records (ADRs) and specifically references internal 'Grove' and 'Amber' SDK abstractions for databases and storage, which are consistent with the author's development ecosystem.- [PROMPT_INJECTION]: The skill exposes a surface for indirect prompt injection because it processes user-provided descriptions of software systems and requirements to generate architectural blueprints.
- Ingestion points: User-provided system requirements, descriptions of component interactions, and feature planning requests (SKILL.md).
- Boundary markers: The skill does not define specific delimiters or instructions to ignore embedded commands within the user's architectural descriptions.
- Capability inventory: The skill results in the generation of text-based context summaries, diagrams, and documentation files within the project's documentation directories (SKILL.md).
- Sanitization: No explicit sanitization or validation of user-provided requirements is specified before they are incorporated into the output blueprints.
Audit Metadata