create-mr
Pass
Audited by Gen Agent Trust Hub on Jun 28, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes various local shell commands using
gitandgh(GitHub CLI) to manage the repository state and create pull requests. These operations include inspecting changes (git status,git diff,git log), pushing branches (git push), and interacting with GitHub's PR system (gh pr create,gh pr view). These actions are core to the skill's intended functionality.\n- [EXTERNAL_DOWNLOADS]: The skill interacts with GitHub, a well-known service, to synchronize code and manage pull requests. These network operations are conducted through official tools and targeting the repository's configured origin remote.\n- [PROMPT_INJECTION]: An indirect prompt injection surface is present due to the processing of untrusted content from the repository.\n - Ingestion points: The skill reads repository-controlled files such as
.github/PULL_REQUEST_TEMPLATE.mdand processes output fromgit diffandgit log.\n - Boundary markers: There are no explicit delimiters or protective instructions used to isolate the ingested repository content from the agent's logic during PR generation.\n
- Capability inventory: The skill has capabilities to transmit data to remote systems via
git pushandgh pr create.\n - Sanitization: No evidence of sanitization or validation of the ingested repository content is found before it is interpolated into the prompts for generating the PR title and body.
Audit Metadata