co-archive
Pass
Audited by Gen Agent Trust Hub on Jun 20, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious code, obfuscated content, or suspicious network operations were identified. The skill's behavior is consistent with its stated purpose of managing change archives within a project directory.
- [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface because it processes content from external
CHANGE.mdfiles during the verification and archiving process. - Ingestion points: The file
openspec/changes/<slug>/CHANGE.mdis read in step 1 of the procedure. - Boundary markers: There are no specific delimiters or instructions to ignore embedded commands within the processed file.
- Capability inventory: The skill triggers the
spec_archivetool, which has the capability to write to and move files within therepo_root. - Sanitization: No sanitization or validation of the text content within
CHANGE.mdis specified before it is processed by the agent or the archive tool.
Audit Metadata