Homeโ€บBlogโ€บAgentic AI System Designs
Agentic AI System Designs ยท Harness teardown

Hermes Agent, Taken Apart

Nous Research's open-source Hermes Agent, dismantled into eight layers: a gateway to six chat platforms, curated memory and session search, skills it writes for itself, subagents, a scheduler and six execution backends.

FL
FrontierAGI Team
โš ๏ธ How to read this (as of 7 Nov 2026)Facts come from the hermes-agent GitHub repository (read 7 Nov 2026). Its documentation site could not be reached from our build environment, so the agent loop, policy and model details are marked โš ๏ธ. The RSI classification is ours. Build times, team sizes and costs are our estimates. Disclosure: Claude, made by Anthropic, helps write this site, and this series covers Anthropic, Google and OpenAI products.

1. Case file

MakerNous Research
CategorySelf-improving harness
LicenceMIT
Runs onLinux, macOS, WSL2, Windows, Android/Termux
Talk to it fromTerminal + 6 chat platforms
GitHub stars~261k when read (7 Nov 2026)

2. The whole system in one picture

๐Ÿ›๏ธ The whole system, layer by layer, with one task traced

โ” control (who calls whom)โ”… data (what comes back)โ–ญ solid border: runs on your machineโ”… dashed border: runs in the vendor's cloud
Press play to follow one task down through the layers and back, or tap any component.

Same eight-layer reference stack in every article of the series. Components and calls are drawn from the project's documentation and code; where internals are not public the drawing is inferred from documented behaviour. The monospace box shows what actually moves between layers at each step (simplified).

3. Layer by layer

L1๐Ÿ–ฅ๏ธ Interfaces ยท Where people (or other agents) talk to it

What it does
Where you reach it.
How it is built here
CLI/TUI plus Telegram, Discord, Slack, WhatsApp, Signal and email; installs on Linux, macOS, WSL2, native Windows and Android/Termux.
Design choice
Meet people where they already talk, not only in an editor.
Without it
The agent is only useful at a desk.

L2๐Ÿงญ Sessions & orchestration ยท Who owns a task: sessions, background tasks, subagents

What it does
Routing, helpers, schedules.
How it is built here
A single gateway process for all platforms; isolated subagents; a built-in cron scheduler.
Design choice
An always-on service rather than a session you open.
Without it
No way to receive messages or run jobs on a schedule.

L3๐Ÿ” Agent loop ยท How it decides the next step

What it does
The decision loop.
How it is built here
Core agent loop โš ๏ธ (documented on a site we could not reach).
Design choice
Unknown in detail; the distinctive work happens in layers 4 and 5.
Without it
โ€”

๐Ÿ” How this loop compares with the other four, dial by dial, and its RSI level: Inside the Agent Loop

L4๐Ÿง  Context engine ยท What the model sees on each turn

What it does
What it remembers.
How it is built here
Agent-curated memory with periodic nudges, FTS5 full-text session search, and Honcho user modelling.
Design choice
Memory curated by the agent itself, plus search over everything it has done.
Without it
Every conversation starts from zero.

L5๐Ÿ› ๏ธ Tools ยท What it can do

What it does
What it can do, and learn to do.
How it is built here
Tools, skills it creates after complex tasks (agentskills.io-compatible), and Python scripts that call its tools over RPC.
Design choice
The agent writes its own procedures: skills become reusable know-how.
Without it
It solves the same problem from scratch every time.

L6๐Ÿ›ก๏ธ Policy & safety ยท What it may do

What it does
What it may do.
How it is built here
Depends on the chosen backend and configuration โš ๏ธ; not detailed in the README.
Design choice
Safety comes largely from where it runs (layer 7).
Without it
A self-improving agent with broad local access is risky.

L7๐Ÿ“ฆ Execution ยท Where actions actually run

What it does
Where commands run.
How it is built here
Six terminal backends: local, Docker, SSH, Modal, Daytona, Vercel Sandbox.
Design choice
Execution is swappable: laptop, container, remote or cloud sandbox.
Without it
One environment for every job, safe or not.

L8๐Ÿงฌ Model & telemetry ยท Which model, and what gets logged

What it does
The model.
How it is built here
A model you configure โš ๏ธ.
Design choice
Model choice left to the user โš ๏ธ.
Without it
โ€”

4. Connecting the dots

๐Ÿงฌ Design DNA: the three decisions that define it

Learn from its own workcurated memory, session search, self-written skills
Live where you chatone gateway, six platforms
Swappable executionsix terminal backends

๐Ÿ’ธ Where the tokens go in the traced task (our estimate)

Memory and skills loaded per task25% ยท ~75k
Session search results10% ยท ~30k
Tool results30% ยท ~90k
Subagent work20% ยท ~60k
Model output15% ยท ~45k

An illustrative split for a ~300k-token task: about $1.20 at an assumed blended $4 per million tokens. Real numbers vary widely with the repo, the model and caching; the shape (context re-sent every turn dominates) is the point.

5. How it evolved

Unknown โš ๏ธ Launch date not stated on the pages we could read.
Nov 2026 MIT-licensed; ~261k stars and ~57.8k forks; features as listed in the README (read 7 Nov 2026).

6. Where it breaks, and what it does about it

FailureLayerWhat happensWhat this system does about it
Bad lessons stickL4Wrong facts saved to memory or skills are reused.Review skills it writes; prune memory โš ๏ธ.
Skill driftL5Self-written skills accumulate or contradict each other.Treat skills like code: version and review them.
Chat-borne injectionL1Anyone who can message it can try to instruct it.Restrict who can reach the gateway; prefer sandboxed backends.
Broad local accessL7Local backend with full user permissions.Use Docker or a cloud sandbox backend.
Unattended jobsL2Cron jobs fail silently or act repeatedly.Log and alert on scheduled runs.

7. Security and threat model

๐Ÿงฑ Trust boundaries

Leaves your machinemessages via third-party chat platforms; context sent to the model; work sent to cloud backends if chosen
Untrusted inputschat messages (anyone who can reach the bot), files, web content, its own past memories
Powerful actionscommands on the chosen backend; scheduled jobs; skills it writes and later runs
Controlsbackend choice (sandboxed vs local), who can reach the gateway, reviewing skills and memory โš ๏ธ

8. Standards it shares, and benchmarks

9. What people use it for

IndividualPersonal assistantMessage it from your phone to run tasks on your machines.
IndividualRecurring jobsDaily checks and reports via cron.
TeamChat-ops agentA Slack or Discord agent the team can ask.
TeamGrowing playbookSkills accumulate into team know-how.
EnterpriseSelf-hosted, openMIT licence, your choice of backend.
Beyond codeResearch and adminSearches, summaries, file chores (typical uses).

10. How to build one from scratch

mini_hermes.py ยท Python 3, standard library only; scripted stub model
"""mini_hermes.py โ€” a toy harness in the style of Hermes Agent (illustrative).

A gateway receives chat messages; the agent searches past sessions (SQLite FTS5,
falling back to a simple scan if FTS5 is missing), reuses a skill it wrote last
time, and writes or updates skills after a task. The "model" is a scripted stub;
standard library only.
"""
import sqlite3
db = sqlite3.connect(":memory:")
try:
    db.execute("CREATE VIRTUAL TABLE sessions USING fts5(text)"); FTS = True
except sqlite3.OperationalError:
    db.execute("CREATE TABLE sessions(text)"); FTS = False
SKILLS = {}

def search(q):
    if FTS: return [r[0] for r in db.execute("SELECT text FROM sessions WHERE sessions MATCH ?", (q,))]
    return [r[0] for r in db.execute("SELECT text FROM sessions") if q in r[0]]

def handle(platform, msg):
    hits = search("date")
    skill = SKILLS.get("fix-date-tests")
    steps = skill["steps"] if skill else ["run tests", "inspect parse()", "try formats", "fix", "re-run"]
    db.execute("INSERT INTO sessions VALUES (?)", (f"{platform}: {msg} -> fixed date parsing",))
    if skill: skill["uses"] += 1
    else: SKILLS["fix-date-tests"] = {"steps": ["run tests with TZ=UTC", "check split format", "fix", "re-run"], "uses": 1}
    return f"[{platform}] recalled {len(hits)} past session(s), {'reused skill' if skill else 'no skill yet'}, {len(steps)} steps"

def main():
    print(handle("slack", "date test failing, please fix"))
    print(handle("telegram", "the date test broke again"))
    print(f"skills: {[(k, v['uses']) for k, v in SKILLS.items()]}  session search: {'fts5' if FTS else 'scan'}")

if __name__ == "__main__":
    main()
$ python3 mini_hermes.py [slack] recalled 0 past session(s), no skill yet, 5 steps [telegram] recalled 1 past session(s), reused skill, 4 steps skills: [('fix-date-tests', 2)] session search: fts5
PhaseWhat you buildTime*Team*Cost*
0 ยท PrototypeLoop with tools and one chat platform1โ€“2 weeks1$15โ€“40k
1 ยท MemorySession store, full-text search, curated memory1โ€“2 months2$100โ€“250k
2 ยท LearningSkill writing and reuse, user model, review tools2โ€“4 months3โ€“5$300โ€“800k
3 ยท ReachGateway for many platforms, scheduler, backends2โ€“4 months3โ€“6$300โ€“900k
4 ยท Community productInstallers for every OS, docs, security hardening6+ months5โ€“15 incl. contributors$1โ€“3M a year

*Our estimates, not vendor figures: fully loaded engineer cost assumed at $250k a year (US-style salaries; lower elsewhere), plus model API and infrastructure. Treat as orders of magnitude.

๐Ÿงฎ What would running your own cost a month?

Our model: model cost = tasks ร— tokens ร— price ร— 30 days; people = engineers ร— $250k รท 12; infrastructure (sandboxes, storage, logs) assumed at 15% of model cost. A blended price mixes cached input, fresh input and output; check your provider's current prices.

11. When to choose it

โœ… Choose it when

  • A personal or team agent you message from anywhere.
  • You want it to get better at recurring tasks.
  • Open source and your choice of execution backend.

โš ๏ธ Think twice when

  • You need strict change control: self-written skills need review.
  • You need enterprise support.

12. How it compares, layer by layer

HarnessL1InterfacesL2Sessions & orchestrationL3Agent loopL4Context engineL5ToolsL6Policy & safetyL7ExecutionL8Model & telemetry
Claude CodeTerminal, IDE, desktop, web, Slack; Agent SDKSessions with resume; subagents; cloud sessions on the webOne general think โ†’ act โ†’ observe loopCLAUDE.md, auto memory, compactionBuilt-in file/shell/search tools, MCP, skillsPermission rules and modes, hooksYour machine (optional sandboxing โš ๏ธ); cloud for web sessionsClaude models; transcripts, cost and usage telemetry
CodexTerminal, IDE, web (chatgpt.com/codex)Parallel cloud tasks; local sessionsAgent loop per taskAGENTS.md, repo contextShell, file edits, MCPApproval modes โš ๏ธOS sandbox locally โš ๏ธ; container per cloud taskOpenAI models; task logs and diffs
Gemini CLI / JulesTerminal (CLI); browser + GitHub (Jules)Interactive session vs asynchronous taskAgent loop (CLI); plan-then-execute (Jules โš ๏ธ)GEMINI.md context filesBuilt-in tools, MCPConfirmations in the CLI; plan approval in Jules โš ๏ธYour machine (CLI); cloud VM (Jules โš ๏ธ)Gemini models (CLI "not optimized for Gemini 4+")
Deep AgentsYour own app (library)LangGraph threads; sub-agentsPlanning loop on LangGraphFilesystem as memory; offloadingYour tools, MCP, skillsHuman-in-the-loop interruptsPluggable backends: local, sandboxed, remoteAny chat model; LangGraph state and tracing
Hermes AgentTerminal + Telegram, Discord, Slack, WhatsApp, Signal, emailGateway, subagents, cron schedulerAgent loop (details on its docs site โš ๏ธ)Curated memory, FTS5 session search, user modelTools, self-written skills (agentskills.io), Python RPC scriptsDepends on backend and configuration โš ๏ธSix backends: local, Docker, SSH, Modal, Daytona, Vercel SandboxModel you configure โš ๏ธ

One line per layer; our reading of each project's documentation and code as of 7 Nov 2026.

HarnessRuns whereOpen sourceSubagentsPersistent memorySkillsMCPSandboxHuman approvalAsync / cloud tasks
Claude CodeTerminal, IDE, desktop, webโ€”โœ“โœ“โœ“โœ“โ—โœ“โœ“
CodexTerminal, IDE, cloudโœ“ (CLI)โ—โ—โ—โœ“โœ“โœ“โœ“
Gemini CLI / JulesTerminal (CLI), cloud VM (Jules)โœ“ (CLI)โ—โ—โ—โœ“โ—โœ“โœ“
Deep AgentsYour Python / JS appโœ“โœ“โ—โœ“โœ“โ—โœ“โ€”
Hermes AgentTerminal + chat appsโœ“โœ“โœ“โœ“โ—โœ“โ—โœ“

โœ“ built in ยท โ— partial or via add-ons ยท โ€” not a feature. Our reading of each project's documentation as of 7 Nov 2026; check before relying on it.

13. The series

Agentic AI System Designs: four kinds of agent system, one teardown each

Start here๐Ÿงญ Four ways to build an agentHarness๐Ÿงฐ Claude CodeHarness๐Ÿงฐ CodexHarness๐Ÿงฐ Gemini CLI, Antigravity & JulesHarness๐Ÿงฐ Deep AgentsHarness๐Ÿงฐ Hermes Agent (this article)Framework๐Ÿงฑ OpenAI Agents SDK ยท comingFramework๐Ÿงฑ Google ADK ยท comingFramework๐Ÿงฑ LangChain / LangGraph ยท comingFramework๐Ÿงฑ CrewAI ยท comingFramework๐Ÿงฑ Claude Agent SDK & Microsoft Agent Framework ยท comingRuntimeโš™๏ธ Amazon Bedrock AgentCore ยท comingRuntimeโš™๏ธ LangSmith Deployment ยท comingRuntimeโš™๏ธ Temporal ยท comingRuntimeโš™๏ธ Vertex AI Agent Engine & Azure AI Foundry ยท comingBuilder๐Ÿงฉ n8n ยท comingBuilder๐Ÿงฉ Zapier Agents ยท comingBuilder๐Ÿงฉ Microsoft Copilot Studio ยท comingBuilder๐Ÿงฉ Make ยท coming

All of them on one page: Agentic Systems at a Glance.

๐Ÿ“– Glossary: 15 terms used across the series
Agent loop
The repeat cycle: the model picks an action, the harness runs it, the result goes back to the model.
Harness
Everything around the model that turns it into a working agent: loop, tools, context, policy, execution.
Context window
The text the model can see at once. Everything the agent knows on a turn has to fit in it.
Context compaction
Summarising older conversation so a long task keeps fitting in the context window.
Tool schema
A machine-readable description of a tool (name, inputs) that tells the model how to call it.
MCP
Model Context Protocol: an open standard for connecting agents to outside tools and data.
Subagent
A second agent with its own context, used for a side task; only its summary comes back.
Skill
A packaged set of instructions (and sometimes scripts) the agent loads when relevant.
Hook
Your own command that runs automatically before or after an agent action.
Sandbox
A restricted place to run commands so mistakes or attacks cannot reach the rest of the system.
Approval mode
The rule for what the agent may do without asking you first.
Prompt injection
Instructions hidden in data the agent reads (a file, a web page) that try to hijack it.
Trace
The full record of a run: every model call, tool call and result.
AGENTS.md / CLAUDE.md / GEMINI.md
Plain markdown files of project instructions that harnesses read at the start of a session.
Asynchronous agent
An agent you hand a task to and leave; it returns a result (often a pull request) later.

14. Sources