The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill configures the agent to ingest data from external sources like GitHub Issues and GitLab Merge Requests, which may contain malicious instructions.
Ingestion points: Commands such as gh issue view and glab mr view are configured to read descriptions and comments from issue trackers as defined in issue-tracker-github.md and issue-tracker-gitlab.md.
Boundary markers: The provided templates lack specific delimiters or instructions for the agent to distinguish between data and instructions within the fetched ticket content.
Capability inventory: The agent following this setup typically possesses capabilities for file modification and command execution.
Sanitization: No explicit sanitization or validation of the retrieved ticket data is prescribed.

setup-matt-pocock-skills