Engineering notes.

Hermes Agent plugins now declare and self-install Python dependencies

Plugin Python deps now auto-install with constraint resolution, conflict refusal, and survival through hermes update. 785 LOC, 17 files, 587 tests passing.

• hermes
• plugins
• python
• dependencies

Hermes Agent Ships Plugin Catalog With 100 Pre-Vetted Extensions

65 curated MCP servers and 96 community plugins, all Nous-reviewed, with interactive install and selective tool enabling.

• hermes
• mcp
• plugins
• catalog

Hermes `tools.stub_mode` reduces non-core tool schemas to stubs

Hermes `tools.stub_mode` reduces non-core tool schemas to stubs, cutting ~2,756 tokens per call. With 25 MCP tools loaded, trajectory compression is deferred by about 15 turns.

• hermes
• tokens
• tool-schema
• context-window

Cutting Hermes Agent overhead with prompt-size and on-demand tools

A recent X thread broke down how to measure Hermes fixed token cost, then reduce it by disabling skills, trimming memory, and switching tool_search to auto.

• hermes
• tokens
• prompt-engineering
• optimization

Hermes Agent Cuts Tool-Definition Tokens by Half With Progressive Disclosure

Hermes Agent's new tool_search defers MCP schemas until needed, cutting tool tokens ~50% while improving accuracy from 49% to 74% on Claude Opus 4.

• hermes
• mcp
• tokens
• optimization

Hermes cron jobs now persist memory between runs

Scheduled agents load persistent memory and skip LLM calls when nothing changed. Cron continuity and monitor mode cut redundant inference on recurring checks.

• hermes
• cron
• memory
• optimization

Hermes Agent Replay Compaction Cuts Re-sent Tool Output by up to 80%

Hermes Agent PR #87303 compacts oversized tool results on send. 80% of replayed input removed (5x), up to 120x per oversized result, 10 providers.

• hermes
• context-optimization
• token-savings
• replay-compaction

Hermes Agent Cut 13,103 Tokens From Skill Activation by Dropping a Duplicated Path

Skill activation in Hermes Agent dropped from 42,145 to 29,042 tokens (-31.1%) on a 462-file skill after a duplicated absolute path was removed from each supporting-file line.

• hermes
• token-optimization
• skills
• context-compression

Hermes Agent cut 282 tokens per call by deleting prompts its memory tool already teaches

Hermes Agent's memory/skills guidance dropped from 537 to 255 tokens per call (-282) by removing instructions the memory tool schema already teaches.

• hermes
• prompt-engineering
• token-efficiency
• system-prompt

Hermes Agent's Clarify Tool Asks Up to Five Questions in One Call

Hermes's clarify tool now asks 2-5 questions per call instead of one, cutting agent-human round-trips and the 16-27 option combinatorial blowup.

• hermes
• agents
• prompt-engineering
• round-trips

Hermes Agents Render Live Inline UI in Chat Via Plugin ::directives

Hermes plugins render live inline UI in the chat transcript. Agents emit ::directive lines, plugins claim the name, output paints in a sandboxed frame.

• hermes
• plugins
• generative-ui
• desktop

Hermes Cron Jobs Now Carry Their Previous Output Into Every Run

continuity=true injects a cron job's own 8K-truncated last output into each run, so recurring jobs dedupe against what they already reported.

• hermes
• cron
• context
• automation

Hermes Agent Adds Opt-In Support for Agent Plugins v1, the First Cross-Vendor Standard for Agent Extensions

Agent Plugins 1.0: a portable format for agent skills and MCP servers, backed by AWS, Cursor, Microsoft, OpenAI, and Vercel. Hermes shipped support same day.

• hermes
• agent-plugins
• mcp
• skills

Hermes X Roundup: August 8, 2026

Fail-closed shell hooks, AGENTS.md directory-chain merging, slash-menu fuzzy scoring, and --resume latest arrive in Hermes. GitHub-sourced (X API paywalled).

• hermes
• roundup
• hooks
• cli

Meta's Muse Harness Cut Agent Token Usage 2.7x by Changing the System Prompt

Cline injected Meta's Muse Spark 1.2 harness principles into their agent prompt. Same model, same task: 19.7M tokens to 7.2M, 49 minutes to 24.

• hermes
• prompt-engineering
• token-optimization
• agent-harness

Hermes Agent X Roundup - August 7, 2026: Slash Command Expansion

/learn, /refine, /heartbeat, and quality gates land in a single 24-hour window alongside STT hardening. All sourced from GitHub - X API paywalled.

• hermes
• slash-commands
• stt
• open-source

Hermes Agent Adds Quality Gates for /goal Commands

Hermes Agent merged three PRs today: quality gates verify /goal output, /heartbeat re-engages idle sessions, and /refine reviews memory on demand.

• hermes
• agent-engineering
• quality-gates
• goal-verification

Hermes Agent Autonomous Lifecycle - /refine, /heartbeat, and Quality Gates Land

Three Prime-Agent features landed: on-demand memory review, recurring idle session prompts, and deterministic goal verification for Hermes Agent.

• hermes
• agent
• automation
• goals

184 Shared Skills Across Two Agents With Zero MCP Servers

zx_joy_ shares 184 Hermes skill files with Claude Code via symlinks. One rule in CLAUDE.md. Bidirectional updates. Zero MCP servers, zero API.

• hermes
• claude-code
• skills
• mcp

Hermes v0.20.0 Ships Voice, Wake Words, Webhooks, and A2A

v0.20.0 adds streaming conversational voice with barge-in, on-device wake words, signed outbound webhooks, grounded citations, and A2A v1.0 agent protocol.

• hermes
• release
• voice
• webhooks

Hermes v0.20.0 - The Herald Release Brings Voice, A2A, and Self-Healing Tools

Hermes v0.20.0: streaming voice with barge-in, wake words, A2A v1.0, self-healing tools, desktop plugin SDK. 3,650 commits, 1,400 PRs, 650+ contributors.

• hermes
• release
• voice
• a2a

System Prompt Optimization Cuts Hermes Agent Fixed Token Overhead by 80%

Hermes Agent system prompts burn 600+ tokens by default. Community optimization cuts that to 120. The skills index alone costs 1.5K-2.5K per turn.

• hermes
• system-prompt
• token-optimization
• skills

Hermes Agent Adds Model-per-Task Control and 120x Model-Picker Speedup

Kanban cards can override model and reasoning effort. The model picker switches from a blocking 7.3s payload build to 0.06s stale-while-revalidate.

• hermes
• roundup
• desktop
• kanban

Hermes Agent Now Caches Tool Schemas on Anthropic - 12K Tokens Stop Getting Re-Sent Every Turn

The 12K-token tool schema was re-sent uncached on every Anthropic call, eating 70% of input spend. A new 4-breakpoint cache plan stops the waste.

• hermes
• anthropic
• prompt-caching
• token-optimization

Tool Schema Caching Lands on Native Anthropic - 70% Spend Cut on Tool-Heavy Agents

PR #76032 re-plans Anthropic's cache budget so the 12K-token tool schema becomes a cached prefix, cutting ~70% of input spend on tool-heavy agents.

• hermes
• prompt-caching
• anthropic
• optimization

The n8n + Hermes Hybrid: Production Agent Deployments With Token-Efficient Tool Design

n8n handles the deterministic backbone; Hermes handles reasoning. Four consolidated webhooks replace 15+ tools, cutting tool-description tokens by 73% per turn.

• hermes
• n8n
• production
• token-optimization

Hermes X Roundup - July 31, 2026

Per-turn micro-compaction lands in Hermes, idle CPU fix saves 20% background burn, FLUX 3 video gen ships, 5 desktop UX PRs land.

• hermes
• roundup
• github
• micro-compaction

Model Routing Plugins Emerge for Hermes Agent After Liquid AI Encoder Release

Two model-routing plugins shipped for Hermes Agent within 48 hours of Liquid AI encoder release. Classifies by sensitivity, role, and difficulty.

• hermes
• model-routing
• plugins
• liquid-ai

Vercel Gateway Returns, Emoji Reactions Land, and Comfy Cloud Joins MCP - Hermes Roundup, July 30

Vercel AI Gateway and Vercel Sandbox return in a 4,259-line PR. iMessage-style emoji reactions land on desktop. Comfy Cloud joins MCP catalog.

• hermes
• roundup
• vercel
• mcp

Crystal Compactor: 93.7% Context Reduction Across a 600-Tool-Call Agent Session

Hermes Crystal compactor reduced 4.17M source tokens to 262K in a 4-hour session — 93.7% reduction with zero continuity breaks, 11ms compactions.

• hermes
• context-compression
• token-optimization
• crystal

Hermes X Roundup: July 29, 2026

A performance sprint on cold-start, voice full-duplex interrupt, server-owned pins, and a 6,316-line CLI decomposition -- sourced from GitHub while X paywalled

oh-my-hermes Layered Memory: Breaking Context Limits With Budgeted Recall

L0-L3 layered memory with per-block 6,000-char budgets prevents context explosion. Combined with a token-disciplined system prompt for max recall per token.

• hermes
• memory
• token-optimization
• oh-my-hermes

Three Perf PRs Land Together: Token Accounting, Wire Client Reuse, Dashboard Query

Three perf PRs target token accounting, HTTP connection reuse, and SQL query optimization. Combined 2,046 additions across 17 files, salvaging community work.

• hermes
• performance
• latency
• token-accounting

Prompt Optimizer Plugin Saves 20-80% Tokens Per Turn With Model-Aware Rewrites

New Hermes plugin rewrites prompts per model family before they hit the LLM, averaging +55 quality points and cutting tokens 20-80% per rewrite.

• hermes
• plugins
• tokens
• prompt-engineering

Runaway-Loop Caps, 60fps Desktop, and Artifacts - Hermes Agent, July 27

Hermes ports Claude Code's session-wide runaway-loop caps, Desktop hits 60fps render, and a public subagent lifecycle API lands for plugin authors.

• hermes
• roundup
• performance
• desktop

Hermes Agent v0.19 Survives Crashes and Resumes Interrupted Tasks Mid-Turn

Hermes v0.19 picks up from the exact turn that crashed, checks what already happened, and finishes the job with no duplication.

• hermes
• reliability
• agents
• infrastructure

Hermes Desktop Lands Four Perf PRs in a Day - Streaming, Timeline, Picker, Multitab

4 desktop perf PRs merged July 26: streaming cost decoupled from transcript length, deferred timeline, faster model picker, and multitab streaming.

• hermes
• desktop
• performance
• roundup

Hermes Iron-Proxy: Your Agent's API Keys Never Touch the Sandbox

Iron-proxy mints opaque stand-in tokens so sandboxed agents never see real credentials. Compromised sandbox = stolen keys are useless.

• hermes
• security
• credential-firewall
• iron-proxy

Hermes Relay Gets Media Parity - Desktop Connects to Remote Gateways, Arabic RTL Lands

Relay Phases 1+2 land media parity (+1,344 LOC), desktop connects to remote gateways, Arabic RTL ships, and session model locking salvages a 3-month-old PR.

• hermes
• relay
• desktop
• i18n

Hermes Graph Routing: Stop Paying a Reasoning Model to Check If It's Tuesday

Monolithic agent calls spend 80% of tokens on steps a bash script could handle. Graph routing with a classifier node yields 5-10x cost savings.

• hermes
• graph-routing
• cost-optimization
• deterministic-paths

Org-Skills Namespace and Context Engine Verbs Land in 48-Hour Window

Hermes Agent merged 51 feature/perf PRs in 48 hours. Org-skills namespace with token-gated discovery pairs with two new ContextEngine plugin lifecycle verbs.

• hermes
• open-source
• roundup

The SILENT Pattern: Cut Hermes Cron Costs by Routing Empty Cycles to Zero Tokens

Most Hermes cron ticks need zero reasoning. The SILENT pattern skips LLM calls on idle cycles. Combined with tiered routing, users report ~60% cost reduction.

• hermes
• cron
• cost-optimization
• silent-pattern

Compression Learns Three New Tricks - Tokens, Time, and Traceability

A compression overhaul: absolute token thresholds, idle-triggered compaction, structured logging. +1,162 lines, 21 files across three PRs by teknium1.

• hermes
• compression
• voice
• kanban

Hermes Auxiliary Models: Stop Paying Opus Prices for Summaries

Hermes Agent has 11 auxiliary model slots for offloading side tasks. Override them with cheap flash models to cut non-reasoning costs by 90%+.

• hermes
• auxiliary-models
• cost-optimization
• configuration

Hermes v0.19.0 Quicksilver - 2,245 Commits, Cross-Surface Themes, and a Secrets Overhaul

Hermes v0.19.0 with 2,245 commits from 450+ contributors. New: cross-surface theme SDK, command secret sources, and per-model compression thresholds.

• hermes
• release
• secrets
• themes

Hermes Agent Quicksilver: 80% Faster Cold Starts, Durable Delivery, and Live Subagent Tracing

v0.19.0 cuts cold-start latency from 4.3s to 0.9s, adds durable response delivery via state.db, and streams subagent transcripts live.

• hermes
• quicksilver
• latency
• durable-delivery

Desktop Perf Sweep Ships 7 PRs in One Day

OutThisLife shipped 7 desktop perf PRs on July 20: virtualized diffs, targeted file-tree revalidation, rAF-coalesced resizes. Teknium1 bundled office skills.

• hermes
• desktop
• performance
• github

Hermes Token Diet: 32% Less Overhead, 96% Less per-Turn Growth

A CLI audits Hermes Agent token burn, applies presets to disable unused tools and skills, cutting fixed overhead 32% and per-turn growth 96%.

• hermes
• tokens
• optimization
• cli

Desktop Performance Wave - 8 PRs Target Rendering Bottlenecks

Hermes Agent main absorbed 52 feature/perf PRs in 48h. @OutThisLife shipped 8 desktop rendering PRs targeting re-render and serialize bottlenecks.

• hermes
• desktop
• performance
• provider-picker

Byte-Stable System Prompts: Hermes Agent Boosts Prompt Cache Hit Rate From 0.1% to 98.7%

Pinning the session-context render and moving per-turn notes to the user message raised Hermes Agent's prompt cache hit rate from 0.1% to 98.7%.

• hermes
• prompt-caching
• performance
• gateway

Hermes prompt cache hit rate jumps from 0.1% to 98.7% in two PRs

Byte-stable system prompts and an api_content sidecar eliminate the first-call cache miss, cutting response time from 27.9s to 2.4-5.8s on 156k-token contexts.

• hermes
• performance
• caching
• gateway

Hermes X Roundup -- July 18, 2026

Terminal billing lands, desktop UX hardens, and Codex gains real-time observability -- all sourced from GitHub while X remains paywalled.

Hermes Agent Lands Per-Task Reasoning Effort Controls - 3 PRs, 35 Files

Three PRs added per-model, per-aux-task, and per-MoA-slot reasoning effort controls to Hermes. Side tasks at low, main model stays at full power.

• hermes
• reasoning
• cost-optimization
• auxiliary-models

Hermes Plugin Interface Gets a Community-Driven Expansion Roadmap

A 4-phase plugin expansion plan lands as a community tracking issue. Kanban gets attachments, desktop rebindable keys, Unreal Engine joins MCP.

• hermes
• plugins
• community
• roundup

Blender MCP Joins the Hermes Agent MCP Catalog With a Curated 4-Tool Surface

Blender MCP joins the Hermes catalog. `hermes mcp install blender` ships a curated 4-tool surface. Three PRs salvaged 620 lines of community reference material.

• hermes
• mcp
• blender
• plugins

Hermes Agent Hackathon Winners: Custodian, Mom-n-Pop Skills, CashFromChaos

NouResearch hackathon winners: Custodian (spending kernel), Mom-n-Pop Skills (small business ops), CashFromChaos (photo-to-sale). 292 submissions, 3 winners.

• hermes
• hackathon
• nous-research
• community

Hermes Agent Lays Out Plugin Interface Expansion Across 18 Sub-Issues

teknium1 files a 4-phase plan for a production-grade Hermes plugin system. API versioning, inter-plugin event bus, pluggable approval, and a community index.

• hermes
• plugins
• architecture
• open-source

Hermes X Roundup: 4x RTX 3090 Server, Content Machines, and Restaurant Booking Skills

4x RTX 3090 Hermes Agent rig doubled client net profit. Plus: Grok 4.5 content machine, local QVAC deployment, restaurant booking API.

• hermes
• roundup
• community

Oh My Hermes: 36 Skills, 7 Agents, One Kanban Board

A community skill collection brings multi-agent Kanban orchestration to Hermes Agent with 36 composable skills and 7 role-specific agents.

• hermes
• multi-agent
• kanban
• skills

Hermes Agent Roundup: Durable Delegations, Codex Usage Reset, DeepInfra Provider

6 feature PRs in 48h: durable background delegations survive restarts, Codex usage reset lands, DeepInfra provider ships, max/ultra reasoning effort levels.

• hermes
• roundup
• github
• delegation

Hermes Gateway Runtime Readiness Checks Replace Binary Liveness

Hermes gateway now reports 8 subsystem checks through /health/detailed. SQLite read-only probe, disk pressure, platform state, queue depth - no SSH required.

• hermes
• gateway
• observability
• health-checks

Hermes Agent Roundup: Desktop Approval, Session Import, and Reasoning Levels

Six feature PRs in 48h: desktop approval mode, session import flow, max/ultra reasoning levels, and CLI workspace binding.

• hermes
• roundup
• github

How Hermes Agent Handles Long Conversations: Dual-Layer Compression and Anthropic Prompt Caching

Hermes Agent uses dual-layer compression and a 4-breakpoint caching strategy to cap context length, cutting tokens by up to 75% on Anthropic.

• hermes
• context-compression
• prompt-caching
• anthropic

Hermes Agent Roundup: max/ultra Reasoning, Workspace Filter, Fireworks AI

July 12 roundup: max/ultra reasoning effort across all surfaces, CLI workspace filtering, gateway readiness checks, and Fireworks AI as a bundled provider.

• hermes
• roundup
• reasoning
• fireworks

Agent Install Monitor Brings Git-Style Auditing to Hermes Environment Changes

Agent Install Monitor records every package install, Docker pull, and git clone executed through Hermes Agent into a local SQLite database for later inspection.

• hermes
• plugins
• devtools
• observability

Hermes Agent Lands on rabbitOS, Desktop Gains Cloud Discovery

Hermes Agent on rabbitOS. Desktop cloud discovery. Community: rabbit r1 voice routing, Grok 4.5 cost analysis, install-monitor plugin.

• hermes
• x-roundup
• rabbitos
• desktop

GPT-5.6, HY3, Clacky: Hermes Agent roundup, July 10

GPT-5.6 lands in Hermes. HY3 free via Nous Portal. Clacky adds Hermes background agents.

• hermes
• gpt-5.6
• hy3
• clacky

Hermes session export unifies across HTML, Markdown, QMD, and HuggingFace

Three PRs merged July 7 deliver 6 export formats, unified filters, and direct HF dataset upload. 3,901 lines of new session portability.

• hermes
• sessions
• export
• cli

GPT-5.6 gets day-zero Hermes support, community pushes 10 slash commands and self-evolving agent taxonomy

Teknium confirms day-zero Hermes Agent support for GPT-5.6. Plus: Shilong Liu's agent taxonomy, IBuzovskyi's 10-command breakdown.

• hermes
• roundup
• gpt-5-6
• open-source

The 25K Token Blind Spot: How Hermes Compression Was Underestimating Context by Half

Post-compression token estimates excluded the tools schema, causing a 20-30K blind spot. The fix was a one-line change.

• hermes
• compression
• token-optimization
• debugging

Hermes X Roundup - July 7, 2026

Buildathon hits SF this week, CuTeDSL skill from Teknium's team, Unbroker skill hits 1,200 bookmarks, and a 4-machine production case study.

• hermes
• roundup
• buildathon
• skills

Hermes Agent Ships Tool Search for MCP: 85% Fewer Tool Tokens, 25-Point Accuracy Gain

Progressive tool disclosure hides deferred MCP schemas behind three bridge tools. BM25 retrieval cuts 22K tokens per turn, Opus 4 accuracy jumps from 49% to 74%.

• hermes
• mcp
• tools
• token-optimization

4 Mac Mini Studios, No Cloud: A Chinese Founder's Hermes Setup

A Chinese founder runs his business on 4 Mac Mini Studios with Hermes Agent, no cloud fees. Plus: AI clone with Venice API, SecretSource commits, Arabic guide.

• hermes
• agent
• local
• mac-studio

Hermes Agent's Record & Replay skill captures macOS workflows via CGEventTap, generates replayable skills

PR #50138 ships 8 files / +4,372 LOC for a record, generate, replay pipeline using CGEventTap. Self-learning mode scores 1.46 vs 3.01.

• hermes
• skills
• computer-use
• record-replay

World Hermes Buildathon: $750K credits, 10 cities, 8 sponsors; Hermes Agent v0.14 cost drops highlighted in the roundup

$750K Hermes Buildathon (10 cities, 5 countries) hit 202L/90B/83K imp. v0.14 cost stories and OpenClaw switchouts lead the day.

• hermes
• x-roundup
• buildathon
• openclaw

Hermes Kanban SQLite WAL Hardening - Six Commits That Stopped Database Corruption Under Concurrent Dispatchers

Eight Kanban SQLite corruption issues fixed in one week. WAL sidecar unlink races, SQLITE_BUSY aborts, torn writes. Six commits harden the surface.

• hermes
• kanban
• sqlite
• wal

Hermes roundup, July 4: production deployments, not feature announcements

Five posts in 24h shifted Hermes talk from hype to production: a studio on four machines, local hardware, gateway search, hermes-coder profile, UNBROKER skill.

• hermes
• community
• deployment
• gateway

Hermes Agent Adds Completion Contracts - /goal Now Judges Done Against Evidence, Not Vibes

Hermes v0.18 /goal gets completion contracts. The judge marks done only when verification evidence is concrete - test output, command result, file excerpt.

• hermes
• goals
• verification
• v0.18

Hermes roundup, July 3: six community builds on v0.18, from a Rust CDP plugin to /bioinformatics

Six builders shipped fresh Hermes v0.18 implementations in 24h: Rust CDP plugin, MCP testimonial, /bioinformatics command, memory how-to, and setup guide.

• hermes
• release
• v0.18
• community

Hermes roundup, July 2: v0.18 Judgement Release ships with 2,333 likes and 626 bookmarks as the day-1 story

v0.18 Judgement: 2,333 likes, 219 RTs, 626 bookmarks, 639K impressions in 24h. Token-cost walkthroughs, offline brains, Nous tool gateway fill the rest.

• hermes
• release
• v0.18
• judgement

Hermes Cron Adds Flat In-Channel Delivery For Slack With Capability-Flag Pattern

PR #56254 adds cron_continuable_surface so Hermes cron delivers flat into Slack channels. Generic capability flag pattern, 13 files, +1115/-1, 280 tests pass.

• hermes
• cron
• slack
• deployment

Hermes roundup, July 1: wandermist's 10x article crosses 350 bookmarks as browser extensions reach 300 stars

Wandermist's 10x-faster article: 121 likes, 352 bookmarks, 141K impressions in 24h. JPIE's browser extension hit 300+ stars in 5 days, 4 releases shipped.

• hermes
• hackathon
• community
• optimization

Hermes Agent Has a Living Map for Memory and Skills - Brooklyn's Shareable Visualization

imbabybrooklyn's living map turns every Hermes agent's accumulated memory and skills into a shareable, animated graph. 85 likes, 39 bookmarks in hours.

• hermes
• memory
• visualization
• community

Hermes Agent Hackathon Final Day - 10x Speed Wins and AgentInvoice, Clawmes Submissions

Hermes Hackathon final day. @NousResearch 328L deadline, @wandermist 10x-speed article, 3 day-one agent submissions (AgentInvoice, Clawmes, Talaria).

• hermes
• hackathon
• community
• optimization

Synapse Brings Graphiti + FalkorDB to Hermes Agent - Salience-Based Decay Over the Endless MEMORY.md

Synapse: Graphiti + FalkorDB Hermes memory plugin with salience-based decay and valid_from/valid_to edges. Low-relevance facts expire, key ones persist.

• hermes
• memory
• graphiti
• falkordb

Hermes Agent Hackathon: NVIDIA x Stripe x Nous - Day-One Submissions from a Multi-Agent Commerce Stack

XONA, MIND Protocol, PIO insurance, Lumina: day-one entries to the Hermes hackathon built on NVIDIA NIM, Stripe, and the agent loop.

• hermes
• hackathon
• nvidia
• stripe

Hermes Agent MoA Presets Ship as Virtual Models - Configurable Reference Fan-Out Across Providers

Hermes Agent ships MoA presets as a virtual model provider with configurable references, aggregator budgets, and provider routing. 30+ commits in 72h.

• hermes
• nousresearch
• moa
• mixture-of-agents

Hermes X Roundup: June 27, 2026 - MoA 2.0 Crosses 4,500 Likes in 17 Hours, mr_r0b0t's /learn Trick Hits 22K Impressions, Hermes One Drops an npm Package

June 27: MoA 2.0 crossed 4,500 likes in 17h, @Teknium's technical breakdown hit 2,000 likes, @mr_r0b0t's /learn trick landed 22K impressions.

• hermes
• roundup
• moa
• mixture-of-agents

Hermes Agent gets a browser surface: three community Chrome extensions land in 66 days

Three independent developers shipped Chrome browser companions for Hermes Agent between April 20 and June 25. All connect to the local API on port 8642.

• hermes
• browser-extension
• chrome
• plugins

Hermes X Roundup: June 26, 2026 - DeepSeek V4 Hits Hermes Agent, Squadic Builds a Financial Flight Recorder, GLM-5.2 Goes Free on Cloudflare

June 26: DeepSeek V4 wired into Hermes Agent (21 likes), Squadic Agentic Money Operator for the hackathon, GLM-5.2 free on Cloudflare Workers AI.

• hermes
• roundup
• deepseek
• hackathon

Content-Addressed Prompt Cache Keys: How Hermes Stops Cron Jobs Going Cache-Cold Every Tick

Hermes swapped cron prompt_cache_key from per-fire session_id to sha256 of the static prefix. Sub-TTL cron fires now reuse the warm prefix.

• hermes
• cron
• caching
• optimization

Hermes X Roundup: June 25, 2026 - /learn Hits the Community, 4,288-Impression Walkthrough, Desktop Perf Fix Lands Same Day

June 25: /learn explainers sweep X, Boxmining's walkthrough clears 4,288 impressions, PR #52273 caps desktop renderer at 20K chars to stop freeze.

• hermes
• roundup
• learn
• slash-commands

Hermes Agent's /learn command turns any source material into a reusable skill

Hermes Agent's /learn turns directories, URLs, conversations, and pasted text into a verified SKILL.md. 11 files, +404 LOC, zero new ingestion engine.

• hermes
• skills
• slash-commands
• agent-design

Hermes X Roundup: June 24, 2026 - Telegram Receipt-Split Bot, Slack-Fronted Agency Stack, /learn for Directories

June 24 roundup: Telegram receipt-split bot, Slack as Hermes frontend, /learn for directories, real estate agent saves $18k/year.

• hermes
• roundup
• telegram
• slack

Hermes Background Review Cut Cost 39x With One Config Field

auxiliary.background_review routes the post-turn review to a cheaper model. 39x cheaper, drops 1 skill and 1 memory on a real 102K session. Prompt-cache aware.

• hermes
• optimization
• cost
• prompt-cache

Hermes lands a stateless one-shot LLM RPC and a cross-platform computer-use preflight

Two PRs in 24h: llm.oneshot gateway RPC bypasses session history (+347 LOC) and a desktop preflight checks cua-driver on macOS, Windows, and Linux (+616 LOC).

• hermes
• agent
• gateway
• computer-use

Hermes iMessage via Photon: gRPC, No Mac Relay, Two Commands to First Text

Hermes v0.17 ships a gRPC-native iMessage channel via Photon. No Mac relay, no public URL. Two commands to first text; persistent stream replaces webhooks.

• hermes
• imessage
• photon
• grpc

Hermes X Roundup: June 22, 2026 - Blank Slate Hits 2,900 Likes, Hermes Unbundled Series Announced

June 22 roundup: Blank Slate mode hit 2,902 likes, manavv announced a Hermes Unbundled three-part series, community debated context compression at 32K tokens.

• hermes
• roundup
• blank-slate
• community

Run Background Subagents in Hermes Agent v0.17.0

delegate_task(background=true) dispatches async subagents. Main agent keeps working. 1,268 lines added.

• hermes
• v017
• subagents
• automation

Hermes Agent Roundup: Desktop Hacks, GPU Farms, and Open-Source Tooling

Sunday's Hermes Agent discussions: Desktop hack videos, OpenClaw-to-Hermes GPU farm migration, Obsidian integration, and a new open-source email tool for your agent.

• hermes
• ai-agents
• desktop
• open-source

Blank Slate, iMessage, and a Fleet of Agents - Hermes v0.17.0 Hits the Community

Hermes v0.17.0 shipped 800 PRs, iMessage via Photon, and blank-slate profiles. Community built fleets, remote setups, and Grok provider hacks within days.

• hermes
• community
• v0170
• blank-slate

Hermes Prompt Optimizer: 55 Quality Points, 20–80% Fewer Tokens

A Hermes Agent plugin rewrites prompts per-model, adding 55 quality points and cutting tokens 20–80% without user effort. SQLite analytics and slash commands for insights.

• hermes
• prompt-engineering
• tokens
• plugins

Hermes Agent as a Platform: A2A, Automation, and Architecture Breakdowns

Hermes Agent is becoming a platform for agent collaboration, automation, and custom workflows. Top posts from the last 24h: A2A plugins, v0.17.0 release (browser control, MCP, Openclaw), and tutorials on building agents from scratch.

• hermes
• ai-agents
• automation
• a2a

Hermes Agent v0.17.0 'The Reach Release' - 800+ PRs, 245 Contributors, Native iMessage & WhatsApp

Hermes Agent v0.17.0 delivers 800+ PRs, 245 contributors, native iMessage/Whatsapp, background workers, and image editing - all open-source and free.

• hermes
• nousresearch
• agents
• open-source

Hermes Agent Roundup: Hardware, Memory Layers, and Real-World Deployments

This week's top Hermes Agent discussions: local hardware stacks, memory layers, unified control planes, and a student's $0/month business built on Hermes Agent.

• hermes
• ai-agents
• self-hosted
• memory

What Kind of Forgetting Breaks Your Workflow? A Framework for Hermes Agent Memory Layers

cobi_bean's framework: pick memory layers by forgetting cost, not plugin availability. Built-in memory, Honcho, and skills each handle a different failure.

• hermes
• memory
• honcho
• agent-architecture

Hermes Curator Stops Auto-Consolidating Skills by Default

The Hermes Agent Curator no longer auto-consolidates. The pipeline cost up to $16/week on frontier models. Capafy launched a Hermes skills marketplace.

• hermes
• curator
• skills
• marketplace

Hermes Agent Can Now Pay for Things - Stripe Link, MPP, and SaaS Provisioning

Hermes Agent adds three optional payments skills via Stripe: virtual cards, HTTP 402 per-request API payments, and SaaS provisioning with spend controls.

• hermes
• stripe
• payments
• mpp

Hermes Agent Gains Traction in China With 40K-Impression Tutorial

A Chinese Hermes Agent tutorial hit 70 likes and 131 bookmarks in 3 days. Community also tackling llm-wiki scaling and corporate Mattermost deployments.

• hermes
• community
• international
• tutorials

Hermes X Roundup: June 16, 2026

June 16 roundup: OpenGradient Veil keeps agent prompts private, AgentBus hits 100 holders, Box positions Hermes as the company brain, Stripe skills reaction continues.

• hermes
• roundup
• privacy
• stripe

Hermes Agent Ships a Bundled LLM Wiki Skill Based on Karpathy's Knowledge Base Pattern

Hermes Agent bundles an LLM Wiki skill that implements Karpathy's interlinked markdown knowledge base. Ingest, query, lint, and Obsidian sync built in.

• hermes
• knowledge-base
• llm-wiki
• karpathy

Hermes Roundup: June 15, 2026

Hermes Agent community roundup - OpenClaw production comparisons, Bangalore builder event, DeepSeek-v4-pro compatibility, Nora agent ops platform.

• hermes
• roundup
• openclaw
• deepseek

Hermes Roundup: June 15, 2026 (PM)

PM roundup: dual agent setups (Hermes VPS + OpenClaw laptop), a practitioner switching from OpenClaw to Hermes, Desktop one-command launch tutorial, and an iOS companion app.

• hermes
• roundup
• openclaw
• desktop

Hermes Agent Automation Blueprints Turn Cron Jobs Into Conversational Workflows

14 parameterized blueprints render natively on every Hermes surface - dashboard forms, slash commands, or conversational fill. No cron syntax required.

• hermes
• automation
• cron
• workflows

Hyre Gateway Adds 11-Model Switching for Hermes Agent

Hyre Gateway published an integration guide for Hermes behind 11-model routing. Cloudzy listed Hermes on its VPS marketplace. Two community guides dropped.

• hermes
• hyre
• gateway
• cloudzy

Teknium's Manim Video Skill Goes Viral on Hermes Agent

Teknium's Manim video generation demo hit 108 likes and 3,100+ impressions. HYRE Gateway shipped 11-model routing. More users migrating from ChatGPT and Claude.

• hermes
• manim
• hyre
• gateway

Hermes Agent Automation Blueprints Turn Cron Into Conversational Workflows

Hermes Agent's new Automation Blueprints turn cron jobs into conversational workflows with three trigger types: Schedule, GitHub Event, and API Call.

• hermes
• automation
• cron
• workflows

Hermes Agent Ships Telegram Rich Messages - Native Markdown Rendering Turns Telegram Into a Lightweight Dashboard

Teknium shipped Telegram Rich Messages for Hermes Agent, enabling markdown output, inline images, and PDF delivery directly in Telegram chat.

• hermes
• telegram
• rich-messages
• nous-research

Hermes Agent Automation Blueprints Turn Cron Jobs Into Conversational Workflows

14 parameterized automation templates with typed slots render natively on every Hermes surface - dashboard forms, CLI slash commands, and agent conversations.

• hermes
• automation
• cron
• blueprints

Hermes Roundup: June 12, 2026

Token Lens plugin, Hermes Desktop app, CREAO AI vs Hermes comparison, Codex self-evolution analysis, and a mobile access PR from the community.

• hermes
• roundup

Headroom Compresses Agent Context by 87% With No Accuracy Loss

Headroom compresses agent context by 87% with no accuracy loss. 60-95% fewer tokens. MCP server gives Hermes Agent single-command integration.

• hermes
• token-compression
• llm
• context-optimization

Hermes X Roundup - June 11, 2026

A complete Hermes Agent guide goes viral in Chinese. Agent profiles, Mystery Shopper skills, and a 4-Mac-Mini local agent farm round out the day's top posts.

• hermes
• x-roundup
• community

Headroom Compresses Agent Context by Up to 95% Without Losing Answers

Headroom compresses tool outputs, logs, files, and RAG chunks before reaching the LLM. 60-95% fewer tokens, same answers. MCP server for Hermes Agent.

• hermes
• context-compression
• tokens
• mcp

Remote Desktop, Commit Velocity, and Multi-Backend Support - Hermes Roundup, June 10

VPS remote desktop workflows drive the week. Five commits land in 24h. Multi-backend support signals an ecosystem maturing past provider lock-in.

• hermes
• roundup
• desktop
• openclaw

Hermes X Roundup: June 10, 2026

Hermes ecosystem roundup: harness-as-moat thesis, ElevenLabs voice integration, model-task-routing skills, retail agent use cases, and in-person meetups.

• hermes
• roundup
• x
• ecosystem

Hermes v0.16 Ships Leaner Skill Defaults - Environment Gating, Built-in Pruning, and Blank-Slate Installs

Environment-gated skills, built-in auto-pruning after 90 days, and blank-slate installs shrink Hermes v0.16's default skill surface.

• hermes
• skills
• optimization
• context-management

gentle-ai Adds First-Class Hermes Integration, Langfuse Monitoring Fixed

gentle-ai v1.37.0 ships full Hermes support. A community PR fixes Langfuse token tracking. Ollama arrives. Desktop GUI gains tool backend management.

• hermes
• gentle-ai
• langfuse
• ollama

Headroom Compresses Agent Context by 60–95% With No Accuracy Loss

Headroom compresses tool outputs, logs, files, and RAG chunks before they reach the LLM. 60-95% fewer tokens with accuracy preserved on GSM8K and TruthfulQA.

• hermes
• context-compression
• agent-optimization
• opensource

When One Model Is Not Enough: Hierarchical Models for Grouped Data

Three models on the same data gave WAIC scores of -2103, -1742, and -1707. The difference: how they handled 30 groups within 600 observations.

• statistics
• bayesian
• hierarchical-models
• multilevel-models

Skill Stress-Testing, Voice Agents, and Multi-Profile Orchestration on Hermes

Community builds: a skill stress-test plugin, a Twilio voice agent, and cheap-model multi-profile orchestration on Hermes.

• hermes
• plugins
• voice
• orchestration

Hermes + Obsidian: The Agentic Second Brain Pattern

Hermes reads Obsidian vaults as live context, creating a second brain that talks back. 2,849 bookmarks on X and $35 hardware builds prove the pattern works.

• hermes
• obsidian
• memory
• knowledge-graph

Hermes X Roundup: June 7, 2026

Weekly X roundup: $4,400/mo from self-hosted Hermes, full agent courses emerge, new Labyrinth observability tool, and OpenClaw migration accelerates.

• hermes
• roundup
• x
• infrastructure

Obsidian + Hermes Agent: A Second Brain That Talks Back

Hermes Agent's Obsidian plugin turns a passive second brain into an active reasoning partner. Notes, backlinks, and wikilinks become live agent context.

• hermes
• obsidian
• knowledge-management
• second-brain

Hermes Agent on X: Memory as Moat, Monetization, and Mid-Turn Steering

Monetization article hits 12.5K impressions, memory layer narrative dominates, mid-turn steering ships as desktop feature, skill PR slashes context 93%.

• hermes
• roundup
• x
• memory

Hermes Agent Ecosystem - June 5, 2026

Multi-profile gateway sockets, RunbookHermes AIOps layer, KongBrain memory for OpenClaw, and Hermes + ElevenLabs phone integration.

• hermes
• openclaw
• ecosystem
• aiops

Hermes Agent Ships Progressive Tool Disclosure - 80 Tools, ~300 Token Overhead

Hermes Agent defers MCP tools behind three bridge tools at 10% context threshold. BM25 retrieval handles 80+ tool catalogs with ~300 token fixed overhead.

• hermes
• tools
• mcp
• context-window

Hermes on X: P2P Payments, Desktop Launch, and 663M Tokens

Warden Protocol adds x402 payments to Hermes for P2P AI inference. Hermes Desktop hits 10K stars. One user logs 663M tokens in 30 days.

• hermes
• roundup
• x

hermesd: A Read-Only TUI Dashboard for Hermes Agent With Zero Write Access

hermesd is a read-only TUI dashboard that monitors Hermes Agent - sessions, tokens, cron, logs - from a single terminal window without writing to agent state.

• hermes
• tui
• dashboard
• observability

Hermes Desktop Lands, Microsoft Ships Scout on OpenClaw

Hermes Desktop public preview for Mac/Windows/Linux. Microsoft ships Scout on OpenClaw. Community builds agent harnesses and skill marketplaces.

• hermes
• openclaw
• microsoft
• desktop

Tool Search Reduces Hermes Agent Context Load by 95.8% Through Progressive Disclosure

Tool Search hides MCP and plugin tools behind BM25 bridges in Hermes Agent. 226-tool catalog drops from 53,994 to 2,289 tokens - a 95.8% reduction.

• hermes
• tool-search
• progressive-disclosure
• bm25

Ara Builds an IDE on Hermes, Desktop v1.0 Ships, Community Calls for Tutorials

Ara launched an IDE powered by Hermes Agent, Desktop v1.0 arrived after 1.8k stars and a month of dogfooding, and the community wants more tutorials.

• hermes
• ara
• desktop
• community

Hermes Agent X Roundup — ComfyUI Design Workflow, Obsidian Memory Stack, Fluxer Plugin

ComfyUI's Hermes Agent design workflow hits 3.1K impressions. Julian Goldie SEO's Hermes+Obsidian memory stack gains 1.9K impressions. Elkim ships Fluxer Discord-like plugin. Memory architecture dominates conversation.

• hermes
• x-roundup
• memory
• plugins

Hermes Skills Load on Demand and Patch Themselves Mid-Run

Tool Search cuts context from 41% to 3%. Skills self-patch on mismatch. Agent capabilities are becoming dynamic and self-maintaining.

• hermes
• tool-search
• skills
• token-optimization

Diagram-as-Code Showdown: 13 Tools, One Pipeline

Same concept, 13 diagram-as-code tools, one API. Mermaid, D2, PlantUML, Graphviz, Excalidraw compared side by side through Kroki.

• diagrams
• kroki
• mermaid
• d2

NVIDIA RTX Spark + Hermes Agent, and Memory Architecture as the Real Differentiator

RTX Spark partnership hits 195 likes. Memory architecture deep dives detail Hermes's layered design: two capped files, frozen prompt, compression flush.

• hermes
• memory
• rtx-spark
• nvidia

Hermes Agent X Roundup — Qwen 3.6 Local, NotebookLM Pipeline, Raycast QuickLink

Hermes Agent runs fully local on Qwen 3.6 27B. NotebookLM + Hermes content pipeline hits 905 impressions. Raycast QuickLink tip surfaces. SkillOS demonstrates on-chain proof of skill with no private key.

• hermes
• x-roundup
• local-ai
• qwen

Hermes v0.15 Shrinks Core by 76%, Adds Multi-Agent Swarm

Hermes v0.15 shrinks the core from 16K to 3.8K lines, adds multi-agent Kanban swarm, and speeds search 4,500x. Community builds cron workflows on top.

• hermes
• release
• velocity
• kanban

Hermes Agent File Writes Are Now Atomic - No Half-Written Files After Crashes

write_file and patch now stream to a temp file, then atomic-rename over the target. A crash mid-write leaves the original intact - no more corrupt files.

• hermes
• file-operations
• atomicity
• crash-safety

Google Ads Gets an MCP Server - The Official Release, the Forks, and the Ecosystem That Followed

Google shipped a read-only MCP server for Ads in early 2026. A fork added write support. Commercial options followed. Here is the landscape.

• mcp
• google-ads
• ai-agents
• advertising

Hermes Ships Tool Search, MCP Optimization, and Infrastructure Hardening in One Wave

Context overhead dropped from 41% to 3% with Tool Search bridge tools; MCP startup went from 7.5s to 115ms; file writes are now atomic.

• hermes
• tool-search
• mcp
• token-optimization

Hermes Agent X Roundup — Tool Search Ships, TUI Dashboard, Step 3.7 Flash Free Access

Tool Search is live and driving a 41%→3% context savings. Korean developer ships local dashboard UI. Step 3.7 Flash free for Hermes Agent users. Community builders shipping plugins.

• hermes
• x-roundup
• tool-search
• agents

Hermes Agent v0.15.0 Adds Built-in MCP Catalog and Skill Bundles

v0.15.0 adds an interactive MCP catalog with one-keystroke install and Skill Bundles loading multiple skills under one slash command. 747 PRs, 321 contributors.

• hermes
• mcp
• agent
• release

The Collapse of Terminal Value Is a Reallocation, Not a Loss

Chamath argues AI makes every moat temporary, collapsing terminal value. The capital doesn't vanish - it flows to startups and assets that don't need moats.

• ai
• markets
• venture
• chamath

Why Your Hermes Cron Job Fails When the Same Prompt Works Manually

Cron jobs run with no memory, no prior messages, and no ambient context. Here is the exact difference and how to write self-contained prompts.

• hermes
• cron
• automation

Hermes Agent Joins Your Discord VC and Talks Back - Here's How

Hermes Agent can join Discord voice channels for live voice conversations, using Whisper STT and ElevenLabs or local TTS. Setup takes three slash commands.

• hermes
• discord
• voice
• tts

Hermes Agent Crosses 90K GitHub Stars, v0.15.0 Ships All 69 Plugin Manifests

Hermes Agent hit 90K GitHub stars as v0.15.0 ships 69 plugin manifests. Step 3.7 Flash lands for agentic workflows. Community roundup.

• hermes
• x-roundup
• open-source
• agents

When Cross-Tokenizer Distillation Breaks Math: X-Token Recovers the 6× Cliff

Tokenizer mismatch drops GSM8k from 12.89 to 2.56 during distillation. X-Token fixes it with a projection matrix and two new loss formulations.

• llm
• distillation
• tokenizers
• nvidia

Hermes Agent Ships Built-In MCP Catalog With Curated Server Registry

Hermes Agent ships a built-in MCP Catalog with curated, pre-vetted server entries. One command installs n8n or Linear with pre-filtered tools and managed OAuth.

• hermes
• mcp
• agent-tools
• nous-research

Hermes on X: Desktop Momentum, Creative MCP, Enterprise Adoption

Hermes Desktop hits 1.6K stars, RunwayML MCP music video, specialist agents, and Dutch enterprises onboarding Hermes in production.

• hermes
• roundup
• x-twitter
• open-source

Hermes Agent Ships Security-Guidance Plugin With 25 Pattern-Matched Vulnerability Rules

Hermes Agent ports Anthropic's security-guidance plugin: 25 rules catch unsafe deserialization, command injection, and XSS at write time. Zero LLM tokens.

• hermes
• security
• plugins
• agent-safety

Cloning Hermes Agent: Every Component, Config, and Cron Job for a Production AI Assistant

Every component, plugin, and cron job in a production Hermes Agent setup, mapped and explained for replication from scratch.

• hermes
• litellm
• deepseek
• skills

Agents Fixing Agents — Hermes Enters Production

Hermes Agent crossed a threshold this week: agents fixing agents, resurrecting decade-old phones, and running production workloads on 1M+ visitor sites.

• hermes
• agents
• openclaw
• production

Hermes Agent Desktop — 20 AI Specialists Collaborate in a Native macOS App

Community desktop app turns Hermes into a 20-specialist multi-agent system with task decomposition, visual Skill Store, and orchestration.

• hermes
• desktop
• multi-agent
• open-source

GBrain + Mnemosyne: Creating the Voltron of Memory Plugins With Hot Cache and Cold Storage

GBrain's 30-page knowledge graph wired into AxDSan's Mnemosyne: a 113-char digest for ambient awareness, and relevance-gated page injection on first turn. Zero MCP calls.

• hermes
• gbrain
• mnemosyne
• memory

Async Media Generation in Hermes Agent With OpenRouter Webhooks

Bridging OpenRouter's async video generation webhooks to Hermes Agent with a Cloudflare Worker signature translator and localhost tunnel.

• hermes
• openrouter
• webhook
• media-generation

GBrain vs Mnemosyne: Architecture, Not Benchmarks

GBrain runs at 250ms per CLI-spawned operation. Mnemosyne runs at 2ms in-process. The difference is architecture, not implementation quality.

• hermes
• memory
• benchmarks
• gbrain

Without Consolidation, Memory Is a Log

Mnemosyne's sleep consolidation cycle scored +70.8pp on BEAM multi-hop reasoning with 0.076ms SQLite reads. Zero cloud dependencies, zero cold start.

• hermes
• memory
• mnemosyne
• sqlite

Deterministic Prefix Ordering: How Production Agents Get 90% Prompt Cache Hit Rates

Prompt caching cuts LLM input costs by up to 90% and TTFT by 80%, but only exact-prefix matches count. How production agents structure prompts for maximum cache hits.

• hermes
• llm
• caching
• optimization

Memory Stacks, Desktops, and OpenClaw Migrations — What X Said About Hermes This Week

A 3-layer memory stack earned 183 likes and 270 bookmarks. Hermes Desktop open-sourced. OpenClaw users migrating. Roundup of Hermes Agent on X.

• hermes
• x
• roundup
• community

Logging Semantic Skill Injection Decisions

Instrumenting the pre_gateway_dispatch hook to log every skill injection decision — score, skip reason, and top-3 matches — so we can tune the threshold with data instead of guessing.

• hermes
• semantic-skills
• observability
• plugins

Semantic Skills — Skill Pre-Loading That Keeps Prompt Caching Intact

semantic-skills v2 pre-loads skills into user messages via pre_gateway_dispatch hook. System prompt stays at 226 tokens, cache-friendly. No first-turn tool call needed.

• hermes
• skills
• caching
• tokens

CodeGraph Slashes Agent Token Burn by 87% Across Our Repos

• hermes
• codegraph
• optimization
• tokens

JIT Skills Slash Hermes System Prompt 95% — Multiplied Across 39 Calls Per Session

• hermes
• tokens
• optimization
• embeddings

How Hermes Agent Builds Its System Prompt

A deep dive into the three-tier system prompt architecture inside Hermes Agent — how it works, why it costs 6,000+ tokens per session, and how we cut that by 95%.

• Hermes Agent
• Prompt Architecture
• Token Optimization
• AI Engineering