Orb setup
OrbCode supercharges Claude Code with Semantic code search, Prompt Optimisation, Code Analytics, and Cost savings.
npx -y skills add MatterAIOrg/cludcode --skill orb-setupAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Run the full OrbCode bootstrap — starts the proxy, authenticates with MatterAI, and kicks off codebase indexing in one shot.
SKILL.md
2.9 KB, 677 tokens by cl100k_base, as published. Nobody here has run it
OrbCode Setup
Run the full OrbCode bootstrap. This will:
- Ask the user what their tech role is (only if not already stored)
- Start the local proxy daemon
- Authenticate with MatterAI (opens browser). Skipped if the user is already logged in.
- Configure
ANTHROPIC_BASE_URLandAPI_TIMEOUT_MSin~/.claude/settings.json - Start the codebase indexer in the background once auth succeeds.
Step 1 — Ask the user's role (BEFORE running setup)
Before running the CLI, check whether the user already has a stored role:
node -e "try { const r = JSON.parse(require('fs').readFileSync(require('path').join(require('os').homedir(), '.claude', 'orbcode', 'role.json'), 'utf-8')); process.stdout.write(r.role || '') } catch { process.stdout.write('') }"
If the command outputs a non-empty string, the user already has a role on file — skip the question and pass --role <stored> through (or omit --role entirely so the CLI reads the file itself).
If empty, ask the user once using AskUserQuestion with this exact configuration:
- question:
"What's your primary tech role? This helps MatterAI tailor optimizations and metrics." - header:
"Tech role" - multiSelect:
false - options (in this order, mutually exclusive — do NOT add an "Other" option, the harness adds one automatically):
- label
"Frontend Engineer", description"UI, web, mobile-web, design systems." - label
"Backend Engineer", description"APIs, services, databases, infra-adjacent code." - label
"Full-stack Engineer", description"Both frontend and backend in roughly equal measure." - label
"DevOps / Platform / SRE", description"Infra, deploy pipelines, observability, reliability."
- label
The user's selected label (or their Other free-text answer) is the role string.
Step 2 — Run setup
If the user passed --token <token> as arguments, pass it through so login runs non-interactively:
node ${CLAUDE_PLUGIN_ROOT}/scripts/orb-cli.js setup --role '<role>' --token '<token>'
Otherwise:
node ${CLAUDE_PLUGIN_ROOT}/scripts/orb-cli.js setup --role '<role>'
If the role was already on file and you skipped the question, you may omit --role — the CLI will use the stored value.
Step 3 — Relay output
Relay the CLI output to the user. After setup completes, tell them:
- Quit and relaunch Claude Code — that's all that's needed for the proxy to take effect
- The
semantic_code_searchMCP tool becomes available once the first batch of files has been embedded
If browser login fails (headless / SSH), the auth URL is printed in the output — direct the user to open it manually, complete login, copy the token, and re-run with /orb-setup --token <token>.
What ships with it
Read from the repository
Just SKILL.md. No reference files, no scripts.