**Competitive pulse - 2026-06-20**

**Bottom line**

Two signals this cycle carry real competitive weight. AWS has GA'd a next-generation OpenSearch Serverless rebuilt explicitly for agentic AI, with 20x faster provisioning, true scale-to-zero, up to 60% cost reduction, and native integrations into Claude Code, Cursor, Codex, Vercel, and Kiro via the OpenSearch Agent Skills protocol. One community commenter said this out loud: "Until now we had to use solutions like Algolia to have true serverless search. Now we can start using OpenSearch for small apps too." That is the displacement framing AWS is actively seeding. Separately, Meilisearch v1.47 shipped personalization on federated search and completed its new settings indexer, continuing its steady infrastructure maturation as a credible open-source alternative with zero operational overhead - a positioning CarJager validated when it migrated from Algolia to cut costs and remove infrastructure burden. The Google AI Visibility story in Search Console is category context, not a direct competitive threat, but it signals that AI attribution is now a real buyer concern in the digital experience market.

**Recommended action**

Owner: Product Marketing

Algolia has MCP. The gap is not capability - it is distribution and proof. AWS OpenSearch Serverless is now shipping agent skill integrations baked into the developer tools your buyers already use daily. Algolia needs visible, documented, working integrations in Claude Code, Cursor, Vercel, and similar environments with public proof: working examples, a maintained opensearch-agent-skills-style repository, and a concrete comparison story on serverless economics. The "scale-to-zero for small apps" narrative AWS is seeding will accumulate in evaluation calls if it goes unanswered. Recommend a focused content sprint: one authoritative Algolia MCP integration guide targeting agentic workloads with reproducible code, published within two weeks.

**Evidence**

- AWS OpenSearch Serverless NextGen GA, May 28: https://aws.amazon.com/about-aws/whats-new/2026/05/amazon-opensearch-serverless-next-generation-generally-available/
- InfoQ technical breakdown with community sentiment, June 8: https://www.infoq.com/news/2026/06/aws-opensearch-serverless/
- OpenSearch Agent Skills protocol (Claude Code, Cursor, Codex integration): https://github.com/opensearch-project/opensearch-agent-skills
- Meilisearch v1.47.0 release notes (personalization on federated search, new settings indexer): https://github.com/meilisearch/meilisearch/releases/tag/v1.47.0
- Meilisearch CarJager case study (Algolia migration, cost reduction, zero DevOps): https://www.meilisearch.com/changelog
- CMSWire on Google AI Visibility in Search Console, June 3: https://www.cmswire.com/digital-experience/google-adds-ai-visibility-reports-to-search-console/

Confidence: high on AWS and Meilisearch signals (primary sources confirmed). The CMSWire / Google signal is category context only - low direct competitive relevance this cycle.

**Watch trigger**

If OpenSearch Agent Skills gains a starred GitHub repository count above 500, or if a second major hosting platform (beyond Vercel and Kiro) ships a native OpenSearch Serverless integration, escalate to Executive Review. Also watch whether Meilisearch publishes a formal MCP server - their roadmap page lists AI Agents and RAG as active product areas.

**Research coverage**

43 sources checked. 31 successful, 12 failed. Signal sources: 4 of 43. Notable failures: Gartner MQ (403 - paywalled), Lucidworks (2 sources failed), Community forums (3 of 4 failed), Algonomy (failed), Perplexity (failed), OpenAI (failed). Tier-1 competitors Bloomreach, Constructor, Coveo, and Google Vertex AI are clean with zero signals this cycle. The forum failure rate (75%) is a coverage gap - community sentiment on OpenSearch and Meilisearch migrations is likely underrepresented.