openspec-archive-change
Pass
Audited by Gen Agent Trust Hub on Aug 31, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to perform directory creation and file movement using shell commands (
mkdir,mv). The<name>variable, which identifies the change to be archived, is interpolated directly into these commands without quotes in the archive step. If a malicious or malformed name is provided by the user or the underlying tool, it could lead to command injection if not properly handled by the agent's execution environment.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data which creates a vulnerability surface for indirect prompt injection.\n - Ingestion points: Reads
tasks.mdfile content and output fromopenspec status.\n - Boundary markers: None present; the agent is not instructed to use delimiters or ignore instructions within these files.\n
- Capability inventory: Executes shell commands (
mkdir,mv) and delegates tasks to a general-purpose subagent.\n - Sanitization: No sanitization or validation of the file content is performed before processing.
Audit Metadata