The Agent Skills Directory

[Indirect Prompt Injection] (LOW): The skill fetches external data from qodo-platform.qodo.ai which is then presented as mandatory instructions for the agent.
Ingestion points: Data is fetched via urllib.request in scripts/fetch-qodo-rules.py and printed to the agent's context.
Boundary markers: While the output uses Markdown headers (e.g., # 📋 Qodo Rules Loaded), it lacks explicit instructions telling the agent to treat the fetched content as untrusted data or to ignore any embedded behavioral overrides.
Capability inventory: The agent is granted Bash and Read permissions and is triggered during high-integrity tasks like code refactoring and bug fixing.
Sanitization: The script performs no sanitization or validation of the rule descriptions before outputting them, allowing potential instructions within rules to influence agent behavior.

get-qodo-rules