The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted content from external sources including pull request titles, descriptions, comments, and Jira workitems. \n- Ingestion points: SKILL.md (PR details, tasks, and Jira items) and references/code-search.md (source code and repository metadata). \n- Boundary markers: The skill lacks instructions for using delimiters or boundary markers to isolate external content from the agent's core instructions. \n- Capability inventory: The agent uses the twg CLI to perform searches and retrieve data across multiple providers (GitHub, Bitbucket). \n- Sanitization: There are no specific guidelines provided for sanitizing or escaping data retrieved from external sources before it is analyzed by the agent. \n- [COMMAND_EXECUTION]: The skill requires the execution of the twg CLI tool. It provides specific instructions for executing the command and includes fallback file paths for the binary on various platforms ($HOME/.local/bin/twg for Unix-like systems and $env:LOCALAPPDATA for Windows).

twg-engineering-work