Config Reference

Note

This page is auto-generated.

Configuration

Schema for the dojo.toml configuration file.

Field	Type	Required	Default	Description
skills	object	No	{"dir":["skills",".agents/skills",".github/skills",".claude/skills",".codex/skills",".gemini/skills",".openclaw/skills",".opencode/skills"]}	Configuration for skill discovery.
model	object	No	{"provider":"anthropic"}	Model provider and model selection for evaluations.
effectiveness	object	No	{"warnFixtureThreshold":4,"confirmFixtureThreshold":12}	Configuration for effectiveness evals.
reporting	object	No	{"perSkill":true,"consolidated":false}	Controls how and where eval reports are written.

skills fields

Field	Type	Required	Default	Description
dir	string | string[]	No	["skills",".agents/skills",".github/skills",".claude/skills",".codex/skills",".gemini/skills",".openclaw/skills",".opencode/skills"]	Directories to search for SKILL.md files. Can be a single path or an array of paths.

model fields

Field	Type	Required	Default	Description
provider	"copilot" | "openai" | "anthropic" | "vercel"	No	"anthropic"	Model provider to use for evaluations. One of: copilot, openai, anthropic, vercel. Defaults to anthropic.
evaluator	string	No	-	Model to use for running evals. Defaults to the provider’s default model. For the vercel provider, use the form ”/” (e.g. “openai/gpt-4o-mini”).
judge	string	No	-	Model to use for judging eval results. Defaults to the provider’s default model.

effectiveness fields

Field	Type	Required	Default	Description
warnFixtureThreshold	number	No	4	Print a warning when a skill has more fixtures than this.
confirmFixtureThreshold	number	No	12	Require —yes confirmation when a skill has more fixtures than this.

reporting fields

Field	Type	Required	Default	Description
perSkill	boolean	No	true	Write a separate report for each skill under its evals/reports/ directory.
consolidated	boolean	No	false	Write a single consolidated report combining all skills.