The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes the ai-todo CLI for all task management operations such as listing, creating, completing, and updating tasks.
[EXTERNAL_DOWNLOADS]: If the CLI is not found, the skill instructs the agent to install the ai-todo-cli package globally using npm install -g ai-todo-cli.
[CREDENTIALS_UNSAFE]: The skill accesses the local configuration file ~/.config/ai-todo/credentials.json to verify the presence of authentication credentials before execution.
[PROMPT_INJECTION]: The skill presents an indirect prompt injection surface as it is designed to automatically ingest and process data from external tool outputs.
Ingestion points: Data is ingested from user-provided task titles, descriptions, git commit messages, and Vercel deployment logs during post-action workflows.
Boundary markers: The prompt instructions do not include specific delimiters or 'ignore embedded instructions' warnings for the data being processed.
Capability inventory: The skill has the capability to execute CLI commands on the local system and interact with a remote API at https://ai-todo.stringzhao.life.
Sanitization: There is no evidence of input validation or sanitization applied to external data before it is passed to the CLI or used to generate task summaries.

ai-todo