The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests raw release notes and GitHub PR data (titles, bodies, and comments) to generate formatted documentation. This data originates from external sources and could potentially contain malicious instructions intended to manipulate the agent's output formatting or content.
Ingestion points: Local file paths provided by the user (e.g., raw-rn.md) and external metadata fetched via the GitHub CLI.
Boundary markers: The instructions lack specific delimiters or "ignore" directives to isolate external content from the system prompt.
Capability inventory: The skill utilizes Read for local file access and Bash(gh *) for network-enabled GitHub repository interaction.
Sanitization: There is no evidence of input validation, sanitization, or escaping of the ingested PR content.
[COMMAND_EXECUTION]: The skill uses the GitHub CLI (gh) via the Bash tool to fetch PR context and repository structure. The use is strictly scoped to gh * commands and targets official vendor repositories.
[EXTERNAL_DOWNLOADS]: The skill is configured to retrieve data from release-notes.kibanateam.dev and the elastic/docs-content repository. These are official resources associated with the skill's authoring organization.

docs-kibana-release-notes