From weave
Reads and updates Weave Claude Code plugin settings including project, API key, agent name, debug mode, and daemon socket. Restarts the daemon to apply changes.
How this skill is triggered — by the user, by Claude, or both
Slash command
/weave:weave-configThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Read and update configuration for the Weave Claude Code plugin.
Read and update configuration for the Weave Claude Code plugin.
If the user invoked this skill with arguments (e.g., /weave:weave-config set weave_project entity/project), execute the corresponding command directly. Otherwise, show the current configuration first and then ask what they want to change.
Run:
weave-claude-code config show
This displays all settings and their sources (settings file vs environment variable).
To update a setting:
weave-claude-code config set KEY VALUE
Writable keys:
| Key | Format | Example |
|---|---|---|
weave_project | entity/project | my-org/my-project |
wandb_api_key | string | abc123... |
agent_name | string | my-team-bot |
debug | true / false | true |
daemon_socket | file path | ~/.weave-claude-code/daemon.sock |
Validation notes:
weave_project must contain a / (entity/project format). Find your entity name at https://wandb.ai.wandb_api_key is available at https://wandb.ai/authorize.agent_name is the name shown for the top-level agent in Weave's Agents view. It must not be empty; surrounding whitespace is trimmed. Defaults to claude-code when unset.WEAVE_PROJECT, WANDB_API_KEY, and WEAVE_AGENT_NAME take precedence over settings file values when set.To read one setting:
weave-claude-code config get KEY
After setting weave_project, wandb_api_key, or agent_name, run weave-claude-code config show to confirm the new value.
The daemon reads these once at startup and persists across Claude Code sessions, so a change is not picked up by a daemon that is already running. Apply it with:
weave-claude-code restart
This stops the running daemon and starts a fresh one. (If none is running, the next Claude Code session starts one with the updated config.)
npx claudepluginhub wandb/weave-claude-code --plugin weaveGuides collaborative design exploration before implementation: explores context, asks clarifying questions, proposes approaches, and writes a design doc for user approval.
Creates structured, bite-sized implementation plans from specs or requirements before writing code. Useful for breaking down multi-step tasks into testable steps with file structure and task boundaries.
Synthesizes the current conversation into a structured spec (PRD) and publishes it to the project issue tracker with a ready-for-agent label, without interviewing the user.