The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses bash and curl to interact with the Dots HTTP API. This is the intended primary purpose of the skill for performing graph operations.
[DATA_EXPOSURE]: The skill manages an API key (sk_... format) stored in a local configuration file dots.json. It provides instructions to the agent to prompt the user for the key and store it securely, including a recommendation to add the file to .gitignore to prevent accidental commits.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data retrieved from the Dots API (e.g., node titles, properties, and graph snapshots). This data could potentially contain malicious instructions intended to influence the agent's behavior.
Ingestion points: API responses from GET /graph, GET /nodes/:id, and /query endpoints defined in SKILL.md.
Boundary markers: The instructions do not specify the use of delimiters or 'ignore embedded instructions' warnings when processing graph data.
Capability inventory: The agent can execute shell commands via curl and bash as defined in SKILL.md.
Sanitization: No explicit sanitization or validation of the graph content is mentioned before the agent processes it.

dots