Hermes Agent README
The self-improving AI agent built by Nous Research.
The Hermes Agent README presents an open agent harness with model-provider switching, terminal and messaging interfaces, scheduled automations, isolated subagents, toolsets, persistent memory, session search, and a closed learning loop around skills.
Classification
- Role
- case-study
- Domain
- cross-domain
- Source type
- doc
- Harness types
- input-shapinggrounding-context-loadingexecution-harnessrepair-harnessmonitoring-harnesslearning-harnesssocial-harnessinterface-harness
- Validation position
- before-generationduring-generationimmediately-after-generationcontinuous
- Validation mode
- mechanicalempiricalsocial
- Prescription stance
- strongly-procedural
- Relation to argument
- capability-is-extendedfirst-mile-input-formationrepairability-mattersobservability-mattersdiffusion-adoption-bottleneck
- Tags
- hermes-agentnous-researchself-improving-agentskillsmemorymessaging-gatewaysubagents
Extended capability commentary
- Input legibility
- Slash commands, personalities, skills, memory, and cross-session search make user intent and prior context available to the model.
- Task structure
- The README describes a full harness surface: CLI/TUI, messaging gateway, scheduler, tools, backends, model providers, subagents, and skills.
- Reward richness
- Hermes emphasizes learning from experience and skill improvement, but the README does not define a single reward signal.
- Feedback latency
- Interactive CLI, messaging interrupts, tool output, scheduled jobs, and session search create frequent feedback opportunities.
- Repairability
- The system can improve skills during use, create skills from experience, search past sessions, and persist knowledge.
- Observability
- Terminal UI, command history, streaming tool output, diagnostics, and session search make agent behavior inspectable.
- Reversibility
- Retry/undo commands are present, but the README does not foreground a broad rollback model.
- Offline evaluability
- Research tooling and batch trajectory generation suggest evaluability, but this is not the main README argument.
- Institutional ratification
- The README is user/harness-oriented rather than focused on organizational approval or governance.
Why it matters
Hermes is an example of the harness conversation moving beyond coding alone: a persistent, multi-surface, model-agnostic agent with memory, skills, automations, and self-improvement loops.
Annotation
The Hermes README is valuable as a productized harness inventory. It does not present a single new model capability. It presents the surrounding system: model-provider switching, a terminal UI, messaging gateways, scheduled automations, persistent memory, skills, subagents, session search, toolsets, terminal backends, and research tooling.
The distinctive claim is the closed learning loop. Hermes says it can create skills from experience, improve skills during use, nudge itself to persist knowledge, search past conversations, and build a user model across sessions. That is a direct capability-extension claim: the agent becomes more useful not only because the model changes, but because the harness accumulates procedural and contextual memory.
Extended Frontier Read
Hermes makes the "agent harness" category concrete across several surfaces:
- interface harness: CLI/TUI plus Telegram, Discord, Slack, WhatsApp, Signal, and email gateway;
- learning harness: skill creation, skill improvement, memory nudges, session search;
- execution harness: local, Docker, SSH, Daytona, Singularity, and Modal terminal backends;
- social harness: cross-platform continuity, user modeling, scheduled reports;
- subagent harness: isolated parallel workstreams and RPC-style tool scripts.
This is not just "a chatbot with tools." It is an attempt to make an agent live where the user lives, remember what matters, and turn repeated work into skills.
Open Questions
- How much of the self-improvement loop is automatic versus user-confirmed?
- Which skills improve reliably during use, and which drift?
- What validation or audit trail exists when memory and user modeling become part of the harness?
Notes
README inspected on GitHub by Codex on Apr 29, 2026. Date is the capture date for this dynamic README snapshot. This entry was prepared with Codex (OpenAI).
Related entries
- What Is an Agent HarnessAparna Dhinakaran · 2026-04-21#skillscapability-is-extendedrepairability-mattersobservability-mattersdiffusion-adoption-bottleneckinput-shapinggrounding-context-loadingexecution-harnessrepair-harnessmonitoring-harnesslearning-harnesssocial-harnessinterface-harness
- LLM Knowledge BasesAndrej Karpathy · 2026-04-01capability-is-extendedfirst-mile-input-formationrepairability-mattersobservability-mattersgrounding-context-loadingexecution-harnessrepair-harnessmonitoring-harnesslearning-harnessinterface-harness
- An open-source spec for Codex orchestration: SymphonyAlex Kotliarskyi, Victor Zhu, and Zach Brock · 2026-04-26capability-is-extendedrepairability-mattersobservability-mattersdiffusion-adoption-bottleneckexecution-harnessrepair-harnessmonitoring-harnesslearning-harnesssocial-harnessinterface-harness
- Resurrecting deceased darlings: The Missing Foreword to AI and the Art of Being HumanAndrew Maynard · 2025-10-18capability-is-extendedfirst-mile-input-formationrepairability-mattersdiffusion-adoption-bottleneckinput-shapingrepair-harnesslearning-harnesssocial-harnessinterface-harness
Overlap is computed on tags, relation-to-argument, and harness types — not on role or domain, because contrasts are often the most useful neighbours.