Skip to main content
The Sovara skill teaches coding assistants how to set up an agent repository, record and inspect runs, rerun LLM steps, annotate examples, and manage lessons. Its Repository Setup workflow is the source of truth used by the guided launcher.

​
Guided repository setup

Run the launcher from the repository that contains the agent:
The CLI:
  1. detects Codex and Claude Code;
  2. asks which assistant to use when both are installed;
  3. globally installs or refreshes the skill for the selected assistant; and
  4. launches that assistant in the current directory with the setup prompt.
The assistant then inspects the repository, installs the matching SDK, adds a project-owned run and useful trace boundaries, and verifies a representative run. It asks before a paid provider call and finishes with an explanation and exact next command. Run it from the directory you want the assistant to modify, or pass that directory explicitly:

​
Guided agent onboarding

After setup, launch a coding assistant to establish human-reviewed ground truth and check the benchmark against it:
The launcher uses the same assistant selection, global skill refresh, and working-directory behavior as sovara setup. The assistant verifies the active Sovara environment and instrumentation, agrees on ground truth with the human and annotates evaluated runs. Lesson changes require human approval, provenance, and retrieval and behavior checks. Use onboard when you have a benchmark or representative sample questions. To improve one existing run without starting the benchmark loop, ask Codex or Claude Code:
A lesson is not always the right fix. The assistant should identify when a tool is wrong or missing, or when permanent guidance belongs in the system prompt.

​
Install the skill without launching setup

Use install-skill for manual, multi-target, or project-local installation: By default, the skill is installed globally. For a project-local install:

​
Useful prompts after setup

​
Refresh the skill