speckit-clarify-zh
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local shell scripts (
.specify/scripts/bash/check-prerequisites.sh) and PowerShell scripts (.specify/scripts/powershell/check-prerequisites.ps1) to verify the environment and identify file paths. While these are intended for prerequisites checking, executing scripts from the local project environment is a sensitive operation.- [INDIRECT_PROMPT_INJECTION]: The skill processes external data from functional specification files to identify ambiguities, which creates a surface for indirect prompt injection. - Ingestion points: The skill loads and scans the file defined by
FEATURE_SPECin Step 2. - Boundary markers: There are no explicit boundary markers or 'ignore' instructions implemented when reading the specification content.
- Capability inventory: The agent has capabilities to execute local scripts (Step 1) and perform file write operations (Step 7).
- Sanitization: The skill includes basic escaping for single quotes in command arguments but does not provide robust sanitization for the content processed from specification files.
Audit Metadata