context-hygiene-for-astra
Pass
Audited by Gen Agent Trust Hub on Sep 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill consists entirely of instructional markdown, licensing information, and configuration metadata. It does not include any scripts, binary executables, or automated commands that could perform unauthorized operations.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze third-party agent instruction files, which is an inherent surface for indirect prompt injection. The skill includes explicit instructions to mitigate this risk. * Ingestion points: The workflow processes SKILL.md, AGENTS.md, and CLAUDE.md files within the agent environment. * Boundary markers: The skill includes a dedicated Boundaries section in SKILL.md that explicitly restricts the agent from performing high-risk actions like payments, deployments, or account changes without separate authorization. * Capability inventory: The skill does not define any subprocess executions, network requests, or direct file-write scripts. * Sanitization: The instructions emphasize using the canonical source of truth and performing a resolution audit to verify the configuration integrity after modifications.
Audit Metadata