The Agent Skills Directory

[COMMAND_EXECUTION]: The skill directs the agent to interact with a local CLI tool named doctrine (including a development build located at ./target/debug/doctrine). These commands are used for searching entities, listing backlog items, and refreshing the project's session state via the boot subcommand. These actions appear to be the primary intended function of the skill within its specific ecosystem.
[INDIRECT_PROMPT_INJECTION]: The skill relies on external data ingested at runtime to make routing decisions, creating a potential surface for indirect prompt injection.
Ingestion points: The agent processes output from the backlog list and doctrine search commands, as well as the content of the @.doctrine/state/boot.md file.
Boundary markers: No explicit delimiters or instructions are provided to the agent to treat this external data as untrusted or to ignore embedded instructions within it.
Capability inventory: The skill allows the execution of the doctrine CLI tool with various arguments to manage project metadata and state.
Sanitization: There is no evidence of sanitization, validation, or escaping of the data returned by the CLI tools before it is processed by the agent.

route