project-discovery

Monorepo structure detection works by reading repository signals like pnpm-workspace.yaml, turbo.json, nx.json, and lerna.json without writing any files. The discovery process maps these markers to the plugin catalog during project classification.

An Evidence template is a deterministic output produced by reading repository signals. It guides downstream project classification and configuration by summarizing detected ecosystems, frameworks, quality gates, and monorepo shape.

Run non-mutating repo discovery during the init step to classify the project. It reads repository signals, identifies ecosystems and quality-gate commands, and outputs an Evidence template to map the project to the plugin's catalog.

Yes, you can apply repo discovery during update steps to refresh owned sections. It reads current repository signals and regenerates the Evidence template to ensure project classification and catalog mapping remain accurate.

Repo discovery detects standard monorepo markers including pnpm-workspace.yaml, turbo.json, nx.json, and lerna.json. It reads these repository signals to determine the monorepo shape and map it to the plugin's catalog.

Non-mutating repo discovery strictly reads repository signals and outputs an Evidence template without writing any files. It does not modify the repository, so downstream tools must handle actual configuration and classification based on the generated evidence.