The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local script ~/paperclip scripts/generate-company-assets.ts with directory names passed as arguments to generate company images. This script belongs to the vendor's local infrastructure and is used for its intended purpose of asset generation.
[INDIRECT_PROMPT_INJECTION]: The skill processes content from multiple COMPANY.md files across the repository to populate the project's root README.
Ingestion points: The agent reads name, description, and source metadata from every COMPANY.md file it discovers during the directory scan.
Boundary markers: There are no explicit instructions to use delimiters or ignore embedded instructions when reading from these external files.
Capability inventory: The skill possesses the ability to modify the root README.md and execute shell commands to generate company assets.
Sanitization: The instructions provide a basic filter by requiring descriptions to be limited to one or two concise sentences, which helps mitigate, but does not eliminate, the risk of malicious instructions in the source files being followed.

readme-updater