The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes shell commands using the Bash tool to perform git operations, run build scripts like gradlew and bazel.cmd, and execute local Kotlin scripts. These operations are core to the skill's purpose of preparing pull requests within the established development environment.
[REMOTE_CODE_EXECUTION]: The skill runs a local Python script suggest_reviewers.py included in its package. This script executes the GitHub CLI (gh) via subprocess.check_output with a safely constructed command list to fetch reviewer metadata.
[DATA_EXFILTRATION]: The skill uses the GitHub CLI (gh) to interact with the JetBrains/intellij-community repository, creating pull requests and querying open PR data. These network operations are intended and scoped to the vendor's repository.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection by processing data from the repository's history and current diffs.
Ingestion points: Reads output from git log and git diff to generate content for PR titles, descriptions, and release notes.
Boundary markers: Does not utilize specific delimiters or instructions to ignore potential commands within the ingested git data.
Capability inventory: The skill has access to tools like Bash and the gh CLI, which provide command execution and network capabilities.
Sanitization: No sanitization or filtering is applied to the data retrieved from the repository's history or file changes.

jewel-pr-preparer