Saturday, November 7, 2026 270 picks · 133 builders credited Subscribe
AgentRags.

The agentic rag, wrung out.

The rag for

Saturday, October 10, 2026

The day in one breath

Nous Research’s $90M Series B at $1.5B valuation and the explicit push into Hermes for Businesses

This is the clear top story. It combines massive traction numbers (24M+ clones, ~2.5% of global tokens), a major funding round with strategic investors (NVIDIA, Microsoft M12, etc.), and a concrete product direction (enterprise features while keeping the open-source core). The simultaneous Hermes Index and free-model expansions show the team is shipping for both builders and future enterprise users at the same time. For the community, it signals that the Hermes ecosystem has crossed from promising open-source project into a well-capitalized platform with long-term staying power. *Curated for high-signal relevance to active Hermes and OpenClaw builders. Sources monitored include official accounts and high-engagement community voices. Deduplication checked against recent days (none prior in this run).*

Industry

Nous Research Series B: $90M at $1.5B, Hermes for Businesses, mobile app

Nous announced a $90M Series B led by Robot Ventures, with Nvidia, Microsoft M12, Samsung, Union Square Ventures, Y Combinator, Menlo Ventures, and others. Valuation is $1.5B. Capital is aimed at Hermes Agent enterprise controls (SSO, permissions, workspace isolation, audit, cost management, Slack/Teams) and a mobile app. Founder note says Hermes has been cloned 24M+ times and, by internal estimate, drives about 2.5% of global token usage. TechCrunch reports ~$36M annualized revenue by mid-September and a company expectation of passing $100M before year-end (WSJ figure, not independently checked here). WSJ, via Yuliya Chernova, also says Hermes overtook OpenClaw and other open-source assistants by usage on OpenRouter.

Credit@NousResearch, @dillonrolnick +1

Releases

Hermes Index Launched — Agentic Performance + Cost Benchmark

Nous launched the Hermes Index, an opinionated leaderboard measuring real agentic task completion and cost-per-task inside Hermes Agent. Claude Opus 5.5 currently leads (~63.31 at ~$4.99/task). Other models (GPT-6 variants, DeepSeek, etc.) ranked with cost transparency. Scores are updated as new models land (e.g., Step 5 Preview adjustments).

Credit@NousResearch

Releases

Hermes Agent v0.21.6 Stable Release + Update Channel Improvements

v0.21.6 shipped (Oct 8) as a stable rollup with the new release pipeline. Recent main merges make stable releases the default for official source installs, add a Desktop channel selector (Stable vs Every commit), and protect against accidental downgrades. Packaged apps (Desktop, Microsoft Store) follow their own cadence.

Credit@BkashJosi

Releases

OpenClaw 2026.10.1 Stable — Stability, Haiku 5.5, Sealed Installs

OpenClaw shipped 2026.10.1 with focus on production stability: sealed immutable installation generations, better Update/Doctor recovery, Claude Haiku 5.5 support, MCP app improvements (keep views open while calling tools), Code Mode persistence, Workboard enhancements, and numerous Windows/Linux/plugin fixes. No major new headline features—emphasis on not breaking working setups.

Credit

Security

High-Severity Hermes Agent Auth Flaw Enables Full Session Takeover — Fixed in v0.21.6

Tenable Research disclosed TRA-2026-65 on October 9. Inconsistent URL parsing (Python urllib vs browser WHATWG) in the native OAuth PKCE flow (`GET /auth/native/authorize`) allowed an attacker to redirect authorization codes to a controlled origin. Combined with attacker-controlled PKCE challenge, this enables full account session takeover. Affects versions before 0.21.6. Fixed in the October 8 release. No confirmed in-the-wild exploitation.

Credit@Python_s_

Never miss a night

Every pick, delivered to your reader.

One feed for everything, or one per agent harness you run. Rebuilt every night, with full credit on every item.