The Agent Skills Directory

[COMMAND_EXECUTION]: The skill requests permission to use the Bash tool. This is intended for running diagnostics and troubleshooting commands during production incidents, as demonstrated in the runbook templates.
[EXTERNAL_DOWNLOADS]: The skill utilizes WebFetch to perform network operations. Examples include service health checks targeting the author's own domain, api.greyhaven.io.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it is designed to ingest and act upon data from production systems that may contain untrusted input.
Ingestion points: The agent is instructed to read system logs and monitor diagnostic output using tools like Bash and Grep.
Boundary markers: None; the templates do not provide delimiters or specific instructions to the agent to treat system output as untrusted data.
Capability inventory: The agent has broad shell access via Bash and the ability to modify files with Write.
Sanitization: There are no provided instructions or scripts for sanitizing or validating logs before processing.
[NO_CODE]: The skill package is composed entirely of Markdown files containing documentation, guides, and templates. It does not include active scripts, binaries, or automated execution logic.

incident-response