spec
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user input via the
$ARGUMENTSvariable and interactive interview responses to populate the generated specification file. - Ingestion points: The
$ARGUMENTSvariable (SKILL.md) and Step 1/3 interactive inputs. - Boundary markers: No specific delimiters or "ignore instructions" markers are used for processing user-supplied data.
- Capability inventory: The skill explores the codebase (Step 2) and writes files to the filesystem (Step 5).
- Sanitization: The slug generation logic (Step 1) performs basic filtering (stripping verbs, removing fillers, joining with hyphens), which serves as a basic sanitization layer for the filename.
- [COMMAND_EXECUTION]: The skill implements logic to derive a file slug and write to
.specs/specs/<slug>.md. While the logic includes character and word filtering, the resulting<slug>is used dynamically in a file-write operation. The risk is mitigated by the fact that the logic only takes the first four words and joins them with hyphens, preventing common path traversal characters.
Audit Metadata