workspace-surface-audit
Pass
Audited by Gen Agent Trust Hub on Apr 6, 2026
Risk Level: SAFE
Full Analysis
- [DATA_EXPOSURE]: The skill instructions specify reading
.envfiles and other configuration files (.mcp.json,package.json) to identify available capabilities. - Evidence: The skill includes a 'Non-Negotiable Rule' stating: 'Never print secret values. Surface only provider names, capability names, file paths, and whether a key or config exists.'
- Context: While
.envfiles are sensitive, the skill is explicitly constrained to only report the presence of specific keys (e.g.,STRIPE_API_KEY) rather than their values, which is consistent with its primary purpose of workspace auditing. - [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the local repository environment, which represents a potential attack surface.
- Ingestion points: The agent reads
package.json,README.md,.mcp.json, and environment files. - Boundary markers: No explicit delimiters are defined in the instructions for separating this data from the system prompt.
- Capability inventory: The skill is primarily read-only and focused on providing analysis and recommendations; it does not perform network operations or execute arbitrary shell commands.
- Sanitization: No explicit sanitization or filtering of the ingested content is described beyond the rule to ignore secret values.
Audit Metadata