spec-driven-cancel
Pass
Audited by Gen Agent Trust Hub on Mar 25, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local Node.js scripts using the
nodecommand. Thescriptsfile contains a relative path (../../dist/scripts) pointing to a directory outside the skill's own folder, which indicates it relies on code from an external location relative to the skill root. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) because it processes data from the filesystem.
- Ingestion points: In
SKILL.md, the agent is instructed to run a command to list active changes from the.spec-driven/changes/directory. - Boundary markers: There are no boundary markers or instructions to ignore embedded commands when processing or displaying the contents of these directories.
- Capability inventory: The skill possesses the capability to execute shell commands (
node) and delete directories based on the ingested data. - Sanitization: No sanitization or validation of the change names (sourced from the filesystem) is specified before they are passed as arguments to the execution command in
SKILL.md.
Audit Metadata