The Agent Skills Directory

Indirect Prompt Injection Surface: The skill processes external data which could contain malicious instructions.
Ingestion points: Fetches issue titles, labels, and metadata from external trackers (GitHub or JIRA) in Step 1.
Boundary markers: Includes explicit instructions for the agent to treat external content as data and never as instructions, specifically referencing an 'injection-guard rule'.
Capability inventory: Performs network queries via gh CLI or JIRA API and executes local Python verification modules.
Sanitization: Instructs the agent to flag potential injection attempts in titles or labels to the user rather than following them.
Command Execution: The skill invokes local command-line tools to perform its tasks.
Verification Scripts: Runs a pre-flight check using python3 -m setup_preflight with a local PYTHONPATH to verify the environment setup.
Tool Invocations: Uses gh auth status and gh issue list to interact with GitHub repositories, which is consistent with its stated purpose as an issue stats dashboard.

backlog-stats