The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute standard development commands such as starting a development server with npm run dev and verifying code with npx tsc. These are routine operations for an agent in a development environment.
[EXTERNAL_DOWNLOADS]: Describes a method for verifying server readiness using curl to poll localhost. This is a localized network operation used for status verification.
[PROMPT_INJECTION]: The skill outlines a process for ingesting external task data from tasks/next_task.md, which creates a surface for indirect prompt injection.
Ingestion points: The agent is directed to read tasks/next_task.md and tasks/progress.txt before starting work.
Boundary markers: There are no explicit instructions to use delimiters or ignore embedded commands within those files.
Capability inventory: The agent has the capability to run shell commands (npm, npx, curl) and discover/load additional skills.
Sanitization: The skill does not prescribe any validation or filtering of the task file content before processing.

agent-workflow-guidelines