commit-message-generator
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from multiple sources including code diffs (via
dxs source changes --with-diffs), branch metadata (commitTitle,commitDescription), and ticket descriptions from CRM and Azure DevOps. - Ingestion points: Data enters the context through
dxsCLI output containing author-provided commit titles, descriptions, and the actual code changes in the branch, as well as external ticket summaries. - Boundary markers: The instructions do not specify the use of clear delimiters (like XML tags or triple quotes) when interpolating this external data into the analysis phase.
- Capability inventory: The skill can execute shell commands via the
dxsCLI, send data to named delivery tools (e.g.,SaveCommitMessage), and write to a knowledge base or Dynamics CRM. - Sanitization: No explicit sanitization or instruction-filtering is mentioned for the ingested diffs or ticket data.
- [COMMAND_EXECUTION]: The skill extensively uses the
dxsCLI tool to perform read and write operations. While these are part of the intended functionality for the Datex Studio environment, the commands incorporate variables like<BRANCH_ID>and<TASK_GUID>derived from user input and previous command outputs.
Audit Metadata