The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the agent to use the bunx package runner to download and execute the @rayhanadev/truffler package from npm. This utility is used to find similar code functions to avoid duplication, but it introduces an external dependency from an unverified third-party source.
[COMMAND_EXECUTION]: The skill executes shell commands using bunx and nr (a package manager runner) to validate code, analyze symbol duplication, and manage repository changesets.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted data such as code diffs and pull request comments.
Ingestion points: Processes external code diffs during implementation reviews and triages human or bot review comments.
Boundary markers: The instructions lack explicit delimiters or safety warnings to ignore embedded instructions within the ingested content.
Capability inventory: The agent has the ability to execute shell commands (bunx, nr), read local file system contents, and generate pull request documentation.
Sanitization: There is no mention of sanitizing or validating the contents of code diffs or comments before they are analyzed or used to generate outputs.

rule-validate