github
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted data from GitHub Issues to drive software implementation tasks, creating a potential vector for indirect instructions.
- Ingestion points: Processes text from GitHub Issues (objectives, background, acceptance criteria) as documented in the 'Issue' section of
SKILL.md. - Boundary markers: Absent; the instructions do not specify a format to delimit Issue content from agent instructions or warn the model to ignore embedded commands.
- Capability inventory: The skill has access to
git commit,git push, and theghCLI for creating and merging Pull Requests as seen inSKILL.md. - Sanitization: Absent; there is no mention of filtering or sanitizing the content retrieved from GitHub.
- [COMMAND_EXECUTION]: The skill leverages shell commands for Git and GitHub lifecycle management.
- Evidence: Explicit instructions for
git commit --author,git fetch --prune, andgh pr mergeare provided inSKILL.md. - Context: These operations are central to the skill's purpose and include defensive measures such as forbidding
git add .and requiring user confirmation for synchronization tasks.
Audit Metadata