The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The spacetime init and spacetime dev commands allow fetching project templates from external GitHub repositories using the --template flag.
[COMMAND_EXECUTION]: The skill facilitates the execution of various spacetime subcommands, including publish, call, sql, and delete, which manipulate database state and configuration.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it processes data from external database logs and SQL query results.
Ingestion points: Output from spacetime logs [DATABASE] and spacetime sql [DATABASE] "<QUERY>" (specified in SKILL.md and references/cli-reference.md).
Boundary markers: None provided; the agent is not instructed to use specific delimiters or to ignore instructions embedded in the tool output.
Capability inventory: The skill provides commands for database state changes and function invocation (spacetime call, spacetime publish, spacetime delete).
Sanitization: No sanitization or validation of the ingested data is specified.
[CREDENTIALS_UNSAFE]: The documentation includes instructions for the spacetime login --token command, enabling manual token entry. Additionally, references/cli-reference-standalone-config.md details how to configure paths for JWT private keys (jwt-priv-key-path) within the config.toml file.

spacetimedb-cli