conventional-commit
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes several git commands to manage the repository state, including
git status,git diff,git log, andgit commit. These commands are standard for version control operations and are executed locally to facilitate the commit process. - [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the local file system through
git diff --cachedto analyze changes. This creates a surface for indirect prompt injection, where an attacker could place instructions inside a file to influence the commit message generation. However, the skill provides a human-in-the-loop mitigation by requiring the agent to present and wait for user approval before executing the commit. - Ingestion points: The agent reads the output of
git diff --cachedinSKILL.mdto identify changes. - Boundary markers: No explicit delimiters are used to wrap the diff content when presented to the LLM.
- Capability inventory: The skill has the ability to execute git commands and perform rebases via the shell.
- Sanitization: The skill relies on user verification and approval of the generated commit message as a primary security control.
Audit Metadata