resolving-merge-conflicts
Pass
Audited by Gen Agent Trust Hub on Aug 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest data from external, potentially untrusted sources such as Pull Request descriptions, issue trackers, and commit messages to determine the intent of code changes. This creates a surface where malicious instructions embedded in these metadata sources could influence the agent's code resolution logic.
- Ingestion points:
SKILL.md(Step 2: commit messages, PRs, issues/tickets). - Boundary markers: Absent; the skill does not define delimiters or provide instructions to treat external content strictly as data.
- Capability inventory: Source code modification via git and local shell command execution for running tests.
- Sanitization: Absent; the agent is not instructed to filter or sanitize the external data.
- [COMMAND_EXECUTION]: The instructions require the agent to discover and execute the project's automated checks (e.g., typecheck, tests, format). This involves running shell commands that could execute arbitrary code if the project's test suite or configuration has been tampered with.
Audit Metadata