The Agent Skills Directory

[COMMAND_EXECUTION]: The skill triggers shell commands including npx skills update -p -y, git status --short, and git diff --stat to manage and report on project updates.
[EXTERNAL_DOWNLOADS]: By using npx skills, the agent initiates a download and execution of the skills package from an external package registry.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted output from external command executions and git diffs, creating a surface for indirect prompt injection.
Ingestion points: The agent reads stdout and stderr from the npx skills update command and the output of git status and diff commands in SKILL.md.
Boundary markers: There are no explicit delimiters or instructions provided to the agent to treat the ingested command output as untrusted or to ignore instructions contained within it.
Capability inventory: The skill utilizes shell execution capabilities via the npx and git binaries.
Sanitization: The instructions do not define any sanitization, filtering, or validation steps for the command output before it is processed by the agent or presented to the user.

update-project-skills