The Agent Skills Directory

Command Execution: The skill automates the creation of GitHub issues using the gh CLI. The skill constructs commands like gh issue create with parameters derived from the agent's session context. While this is a powerful capability, the skill mitigates risk by requiring explicit user confirmation before any command is executed or issue is submitted.
Indirect Prompt Injection Surface: The skill ingests and processes potentially untrusted data to generate bug reports.
Ingestion points: Technical details such as container logs (docker logs), failing command outputs, and the user's initial prompts are gathered as evidence for the report (documented in SKILL.md).
Boundary markers: The skill does not currently specify the use of clear delimiters or instructions to the agent to ignore embedded commands within the ingested text.
Capability inventory: The skill possesses the capability to write to local files and execute network-linked commands via the GitHub CLI.
Sanitization: The skill includes a dedicated redaction step (Step 3) to strip sensitive credentials, passwords, and tokens from all fields before processing, which represents a security consideration regarding data handling.

azuresql-db-feedback