spec-driven-spec-content
Pass
Audited by Gen Agent Trust Hub on Mar 27, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local shell commands including
lsto verify directory existence andnodeto run a specialized project management script (spec-driven.js).- [COMMAND_EXECUTION]: The execution path for the management script is determined via a relative path reference (../../dist/scripts) contained in a local file, pointing to logic located outside the immediate skill directory.- [DATA_EXPOSURE]: The skill accesses local project metadata including.spec-driven/config.yamland.spec-driven/specs/INDEX.mdto retrieve context for routing decisions.- [PROMPT_INJECTION]: The skill has an indirect prompt injection surface because it reads and processes data from external specification files and project indices to determine its routing logic. - Ingestion points:
.spec-driven/specs/INDEX.md,.spec-driven/config.yaml, and existing delta spec files. - Boundary markers: None explicitly defined in the instructions to separate external data from system instructions.
- Capability inventory: Local file reads, directory listing (
ls), and Node.js script execution (node). - Sanitization: No explicit sanitization or validation of the ingested file content is described in the prompt logic.
Audit Metadata