The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill references and suggests fetching documentation and supplementary skill definitions from the vendor's official GitHub repository (github.com/reticlehq/reticle) and documentation site (docs.reticle.sh).
[COMMAND_EXECUTION]: Provides an initialization command npx @reticlehq/server@latest init to set up the Reticle environment. The package originates from the skill author's official namespace on the NPM registry.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and verify application state, console logs, and network traffic to validate test expectations. While this exposes the agent to external data, it is a core functional requirement of the testing framework, and no specific exploits were detected.
Ingestion points: Tool outputs from reticle_act_and_wait, reticle_state, and reticle_run (SKILL.md).
Boundary markers: None specific to the skill body beyond standard tool call structures.
Capability inventory: The skill primarily drives a testing agent but facilitates command execution for server initialization (SKILL.md).
Sanitization: Not explicitly defined in the prompt instructions; relies on the underlying tool implementation for data handling.

agentic-tdd