reach cluster

Partition skill catalogs into cohesive subagent scopes to prevent routing decay.

---

Synopsis

reach cluster [SKILLS] [OPTIONS]

---

Key Scenarios

Partition skills into subagent candidate catalogsTarget subagent catalog sizeExport clusters to JSON or CSV

Partition a skill corpus into high-modularity clusters:

reach cluster ./skills

Specify soft upper bounds on cluster size to create modular subagent workspaces:

reach cluster ./skills --target-size 10

Export cluster assignments for automated subagent scaffolding:

reach cluster ./skills --format json > clusters.json

---

Options

Clustering Options

Option	Type	Default	Description
[SKILLS], --skills	Path	.	Path to the skill directory or corpus to partition (discovered if omitted).
--resolution	Float	1.0	Resolution parameter: higher values yield smaller clusters.
--target-size	Integer	-	Target maximum skills per cluster.
--max-clusters	Integer	-	Maximum number of clusters.
--agent	Choice	-	Agent runtime to query for installed skill locations.
--global, -g	Flag	false	Discover and inspect skills from user global configuration (~/).
--format	Choice	text	Output format: text, json, csv.
--config, -c	Path	-	Path to reach.toml configuration file.

Agent Registry Options

Option	Type	Default	Description
--project, -p	String	None	Google Cloud project ID hosting the Agent Registry.
--location	String	"global"	Agent Registry location endpoint (global, us, eu).
--publisher	String	None	Filter registry skills by publisher identifier (e.g. cloud.google.com).
--registry	Flag	false	Target Google Cloud Agent Registry instead of local workspace.
--fresh	Flag	false	Bypass cached metadata and re-fetch latest skill definitions.
--no-cache	Flag	false	Run without reading or writing local disk cache.