The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill fetches technical documentation from https://docs.battlechain.com/llms-full.txt and installs the cyfrin/battlechain-lib library. These resources are hosted on the vendor's official domains and repositories, which is standard for a development-focused skill.
[COMMAND_EXECUTION]: Provides instructions for using standard blockchain development tools like forge and just. These commands are used for legitimate tasks such as deploying contracts, running scripts, and verifying source code on block explorers.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest external documentation at runtime to provide the agent with current protocol technical details. This creates a surface for indirect prompt injection if the remote content were compromised, though the source is the vendor's official documentation.
Ingestion points: Documentation is fetched from docs.battlechain.com/llms-full.txt as specified in SKILL.md.
Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded commands within the fetched documentation.
Capability inventory: The skill includes the capability to execute shell commands via forge and just (found in SKILL.md).
Sanitization: There are no explicit sanitization or validation steps described for the remote data before it enters the agent's context.

battlechain