The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill recommends installing official vendor tools, including the Terra API CLI via npm install -g @tryterra/cli and brew install tryterra/tap/terra. It also references documentation on the tryterra.co domain.
[INDIRECT_PROMPT_INJECTION]: The skill describes a workflow for fetching and processing wellness data from the Terra API. While this involves ingesting external data, the skill provides specific instructions for handling response objects which helps constrain how the agent interprets the output.
Ingestion points: API responses from api.tryterra.co as described in SKILL.md.
Boundary markers: The skill provides logical structures for parsing specific keys (unsupported, no_data, window) to help the agent interpret data correctly.
Capability inventory: Network operations via curl to the vendor's API endpoints.
Sanitization: The skill includes instructions to not present wellness indicators as medical or clinical data.
[CREDENTIALS_UNSAFE]: The skill uses appropriate placeholders such as YOUR_API_KEY and YOUR_DEV_ID in example API calls, avoiding the exposure of real credentials.

terra-models