fusion-issue-solving
Pass
Audited by Gen Agent Trust Hub on May 9, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes potentially untrusted data from GitHub issue bodies, labels, and discussions, creating a surface for indirect prompt injection.
- Ingestion points: Accesses and processes GitHub issue content and discussions (SKILL.md, Step 2).
- Boundary markers: No explicit delimiters or instructions are provided to the agent to isolate external content from internal instructions.
- Capability inventory: The skill possesses the capability to modify local files, execute shell-based validation commands, and perform remote mutations via the GitHub API (SKILL.md, Steps 7-10).
- Sanitization: No explicit sanitization or filtering logic is defined for the external data ingested from GitHub.
- [COMMAND_EXECUTION]: The workflow requires the agent to execute repository-specific validation and project checks.
- The instructions mandate running "targeted checks" and "required project checks" within the local environment (SKILL.md, Step 8).
- This risk is managed by the instruction that repository-local workflow rules take precedence and that destructive commands require explicit user confirmation.
Audit Metadata