The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill provides instructions for the initial setup of the environment using bun install -g @wordbricks/velen. This is a standard installation procedure for the vendor's command-line utility.
[REMOTE_CODE_EXECUTION]: The skill includes functionality to update its own instructions and the underlying CLI tool via velen update and velen skill update, which fetch content from the wordbricks/skills repository.
[COMMAND_EXECUTION]: The primary function of the skill is executing shell commands via the velen binary. It includes extensive guardrails such as requiring --dry-run for operations that modify external state, providing explicit read-only SQL patterns, and using output bounding flags like --max-rows and --max-bytes to prevent terminal flooding.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external sources, including Velen insights and recalled Knowledge Graph memory. It addresses this attack surface by instructing the agent to treat recalled memory as advisory evidence, verify event semantics against local code or source metadata, and use human-in-the-loop confirmations for any downstream mutations.

velen-cli