The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to read and process local project manifest files (package.json) and lockfiles, which are untrusted data sources that could contain malicious instructions.
Ingestion points: The skill analyzes project manifests and lockfiles (SKILL.md, steps 1 and 3) to identify package versions and workspaces.
Boundary markers: No delimiters or instructions are present to prevent the agent from obeying instructions embedded within the manifest files.
Capability inventory: The skill has the capability to execute shell commands via npx and perform network lookups via npm.
Sanitization: No sanitization or validation of the manifest content is described before the data is used to inform agent actions.
[COMMAND_EXECUTION]: The skill executes shell commands using npx and npm to upgrade packages and verify versions.
Evidence: npx remotion upgrade, npm view remotion version, and npx skills update commands are used to perform the core functions of the skill.
[EXTERNAL_DOWNLOADS]: The skill triggers the download of CLI tools and packages from the npm registry using npx.
Evidence: The use of npx remotion upgrade and npx skills update involves fetching and running code from external sources to perform the upgrade process.

remotion-upgrade