The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local bash scripts and system utilities to perform its workflow. It runs ./scripts/analyze_issue.sh to fetch GitHub data and ./scripts/detect_versions.sh to identify system and application versions. It relies on standard tools such as gh (GitHub CLI), jq (JSON processor), uname, sw_vers, and powershell.exe. Additionally, it uses pbcopy, xclip, or clip to provide verification templates to the user's clipboard.
[EXTERNAL_DOWNLOADS]: The skill retrieves data from external GitHub repositories using the gh CLI. It fetches issue details, comments, and PR data from the posit-dev/positron repository. These operations are restricted to a well-known service and a repository that matches the author's identity.
[PROMPT_INJECTION]: The skill has a surface for indirect prompt injection because it processes untrusted data from GitHub. * Ingestion points: Fetches content from GitHub issue bodies, comments, and PR descriptions using the gh tool. * Boundary markers: The skill does not use specific delimiters or protective instructions when passing retrieved content into the LLM context for analysis. * Capability inventory: The skill is capable of writing files to the local filesystem (output directory), accessing the system clipboard, and executing shell commands. * Sanitization: No explicit sanitization or filtering of the fetched GitHub content is performed before the LLM extracts test scenarios and edge cases from it.

positron-qa-verify