FAQ
General
Section titled “General”What is Optra Prism?
Section titled “What is Optra Prism?”Optra Prism is a free intelligence layer for AI-assisted coding. It plugs into Claude Code and gives you visibility into how you use AI — then helps you improve through PRISM scoring, real-time coaching, and usage analytics.
Is it free?
Section titled “Is it free?”Yes. The Personal tier is free forever — no trial, no credit card. It includes full PRISM scoring, coaching, waste detection, and analytics. Team and Enterprise plans add governance features.
What data does it collect?
Section titled “What data does it collect?”- OTEL telemetry — logs, metrics, traces (token counts, timing, model usage)
- Prompt text — your actual prompts, used for PRISM scoring
Prompt text is used for scoring and coaching only. See the privacy policy for details.
Does it slow down my coding?
Section titled “Does it slow down my coding?”No. The plugin hooks are lightweight:
- Prompt scoring: <100ms (local heuristics)
- Response timing: negligible overhead
- Telemetry export: async, non-blocking
Gateway routing adds a small network hop (~50–100ms) if enabled. Disable it for zero latency impact.
Plugin
Section titled “Plugin”Do I need to restart Claude Code after setup?
Section titled “Do I need to restart Claude Code after setup?”Yes. The plugin sets environment variables on session start, so a restart is needed after initial setup or gateway toggle.
Can Prism block my prompts?
Section titled “Can Prism block my prompts?”No. The prompt advisor always exits with code 0. Suggestions are advisory only — your prompt always goes through.
How do I adjust coaching sensitivity?
Section titled “How do I adjust coaching sensitivity?”Change prismThreshold in ~/.prism/config.json. Default is 4. Higher = more coaching. Lower = less coaching.
What if I don’t have an API key?
Section titled “What if I don’t have an API key?”Sign up at dashboard.prism.optra-ai.com to get a free gck_* key.
Scoring
Section titled “Scoring”How accurate is the PRISM score?
Section titled “How accurate is the PRISM score?”The LLM scorer (primary) is ~90% accurate. The heuristic fallback is ~70% accurate. Scores are calibrated against expert-rated sessions.
What’s a good PRISM score?
Section titled “What’s a good PRISM score?”- 7.0+ is Expert tier — strong habits
- 5.0–6.9 is Proficient — solid foundation
- Most developers start at 3.0–5.0 and improve to 6.0+ within 2 weeks
Do scores affect my Claude Code experience?
Section titled “Do scores affect my Claude Code experience?”No. Scores are informational. They don’t limit your access, change model behavior, or affect billing. They’re purely for your awareness and improvement.
How often are sessions scored?
Section titled “How often are sessions scored?”Sessions are scored by the background scoring worker after each S3 write batch (30-second debounce). Scores typically appear in the dashboard within 1–5 minutes.
Dashboard
Section titled “Dashboard”Why don’t I see any data?
Section titled “Why don’t I see any data?”Common causes:
- Plugin not configured — run
/prism:setup gck_YOUR_KEY - Session too recent — wait 1–5 minutes for data processing
- API key issue — verify at
/prism:status
Can my team see my prompts?
Section titled “Can my team see my prompts?”Team leads can see aggregate analytics (scores, usage, waste patterns) but not individual prompt text unless configured by an admin.
Self-hosting
Section titled “Self-hosting”Can I self-host Optra Prism?
Section titled “Can I self-host Optra Prism?”Yes. The entire stack (dashboard, ingest, engine) can run locally or on your infrastructure. See the Architecture page for the component overview and the Environment Variables page for configuration.
What infrastructure do I need?
Section titled “What infrastructure do I need?”- PostgreSQL (Supabase or self-managed)
- NATS JetStream
- S3-compatible storage
- Node.js (dashboard)
- Rust runtime (ingest, engine)