| Core use case |
Operational AI agents (branch/loop/memory/HITL), governance, multi-channel deployment. |
Chat-driven IDE to build JS/Node websites & web apps; fast publish with hosting & domains.
|
| Tech stack |
Model-agnostic with BYOK; REST API + SDKs; gateway overlay. |
Front-end frameworks + Node.js backend only (Python/PHP not supported); built-in hosting, domains,
SSL,
SEO prerender ("SEO Boost"). |
| Build UX |
Visual Agent Studio canvas, prompt library, versioning, low-code flow composition. |
“Build” vs “Discussion” modes, prompt library, file uploads with focus/ignore, version history, GitHub
backup/auto-commits. |
| Data + RAG |
Connectors to SaaS & databases; managed ingestion & indexing; scheduling; dense + sparse
retrieval; SQL indexing. |
Upload files to inform the model; not a managed ingestion/indexing pipeline. |
| Tools & Integrations |
100+ enterprise connectors, custom tools, chaining & scheduling. |
Guides/integrations for Figma, Supabase, Stripe, Netlify, Expo, GitHub; recommended Supabase stack.
|
| Security & Governance |
DLP (100+ detectors), Responsible-AI filters, attack prevention, central policy enforcement &
violation feeds. |
Team prompts/visibility controls; no enterprise DLP/RA policy engine documented. |
| Observability |
Feeds for executions, conversations, tokens, and violations; filters, export, alerts. |
Project history/restore; token guidance; lacks conversation-level audit trails. |
| Access & Identity |
SSO (OIDC) + SCIM 2.0 provisioning; RBAC. |
Teams & GitHub integration; org-level GitHub support appears inconsistent across docs. |
| Data residency & privacy |
Region selection, retention settings, BYOK, clear handling of external provider regions. |
No documented residency posture; provider/platform-controlled processing. |
| Deployment |
API, Web Chat Widget, Catalog, Scheduled Jobs, Email triggers, iOS app/SDK. |
Built-in hosting (free *.bolt.host), domains, quotas, pay-as-you-go overages, SEO
prerender
(“SEO Boost”). |
| Mobile |
Airia Chat iOS app + Swift SDK; Android via REST. |
Ship to mobile via Expo; IDE is desktop-browser focused (no native mobile SDK). |
| LLM policy & tokens |
Per-agent/model governance; token tracking/limits; BYOK cost tracking. |
Underlying Claude Sonnet 4; 200k (free) / 500k (paid) context; “Discussion” mode uses ~90% fewer
tokens;
plan-based rollover/reload rules. |
| Notable campaigns / notes |
Agent-to-Agent protocol; Microsoft AppSource; compliance focus. |
“World’s Largest Hackathon” series; 130k+ participants; >$1M in prizes; active Reddit/Discord
community. |
| Community |
Growing builder content and webinars. |
High-engagement tutorials (e.g., YouTube analytics tool), “10 Bold Bolt Builds,” complete Bolt course
content. |
| Multi-provider |
Multiple models/providers inside one agent; one agent endpoint. |
N/A (single-provider IDE; multi-provider orchestration not the focus). |
| Rapid Retrofit |
AI Gateway: change base URL + key to overlay governance & logging. |
Keep existing Bolt-built UI; route LLM calls through Airia’s AI Gateway for observability & DLP.
|