spec-merger
Pass
Audited by Gen Agent Trust Hub on Aug 5, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill leverages a command-line tool named
ssfto perform state checks (ssf state get) and synchronization operations (ssf sync). These commands are used as intended for the skill's purpose of managing specification workflows and do not involve shell injection or unauthorized command execution. - [DATA_EXFILTRATION]: No network operations, external data transfers, or access to sensitive credentials (such as environment variables or SSH keys) were detected in the instructions.
- [PROMPT_INJECTION]: The skill processes user-provided markdown files in the
changes/directory. While this represents a potential surface for indirect prompt injection, the instructions enforce strict structural parsing of specific headers (ADDED, MODIFIED, REMOVED, RENAMED) and require validation of all candidates before they are published to the main baseline, providing an effective barrier against accidental instruction obedience.
Audit Metadata