Welcome back, Operator
7 missions on file · 9 items need attention
Pick a real Hermes-style capability to test
Run Browser QA with Playwright evidence, then approve the report and generate a receipt.
Use the Codebase Debugger crew with governed terminal_exec calls.
Test the provider-backed web_search tool before giving it to agents.
Use a vision-capable model to describe an image URL and extract findings.
Try image generation or text-to-speech when provider keys are configured.
Drop a Slack/Discord/email-style message into the inbox through the generic webhook.
The control room for the Hermes Agent.
Hermes Agent is Nous Research's open-source autonomous agent — it lives on your server, remembers what it learns, and gets more capable the longer it runs. Hermes Loop is the operator surface: you launch missions, run named crews of subagents, gate risky outputs with approvals, and produce a hashed receipt that proves what happened.
The autonomous engine: persistent memory, subagents with their own conversations and terminals, natural-language cron, and native tools (web search, browser automation, vision, multi-model reasoning).
The governance layer: crews, a job queue, approvals, hashed receipts, a trust ledger, evals, schema self-correction, real-cost accounting, and a full audit trail.
A chatbot gives an answer. The Hermes Agent does the work autonomously; Hermes Loop proves it: who ran, which tools were called, which memory was used, what was approved, and what hash signed off the run.
Run Bug Hunter on /demo-target, approve the QA report, generate the receipt, then open Trust to see the run roll up into risk and proof.
Bug Hunter crew + Playwright browser audit. Real screenshots, real findings.
Codebase Debugger crew using terminal_exec under policy.
Tavily / Brave / SerpAPI sandbox. Provider-gated.
Multimodal vision_analyze. URL or file upload.
Fal / Replicate / OpenAI-compat image generation. Approval-gated.
Natural-language cron parsed into structured cadence.
POST /api/integrations/webhook → Inbox → triage → mission.
What's shipped, partial, or still on the Hermes Agent roadmap.
13 stops from health probe to deployment env vars — live status.
What people can actually do with it
Bug Hunter opens the page in Playwright, captures evidence, finds accessibility/conversion issues, and stages a QA report for approval.
Codebase Debugger lets Hermes use governed terminal commands, then writes a diagnosis without editing files.
Provider-backed web search, vision, image generation, and TTS are testable from the Media section before agents use them.
Use natural-language schedules or a signed webhook to turn external requests into queued agent missions.
Define agents, roles, instructions, tools, and execution order, then run that crew like a built-in workflow.
Receipts and Trust show the timeline, tools, approvals, memory, risk, model cost source, and integrity hash.
- 1 queued · 0 runningOpen queue →Waiting for the worker process to pick up. Make sure `npm run worker` is running.
- 4 approvals pendingReview →Drafts, trade tickets, exports, gated tools.
- 3 memory suggestionsOpen vault →Inferred preferences awaiting accept / reject.
- 2 memory hygiene flagsRun hygiene →Duplicates, conflicts, stale or low-confidence.
Governed terminal_exec + python_rpc for diagnostic crews. Risky commands wait for your approval; safe diagnostics run unapproved.
- [demo] Watch the AI semis space (NVDA, AVGO, AMD) and brief me every weekday mor…triaged
- [mobile] Set up a weekly Monday QA audit for my landing page.new
- [mobile] Audit https://demo.acme-saas.test for conversion + a11y issues.triaged
- [demo] Audit https://demo.acme-saas.test for conversion + accessibility issues.triaged
- RX-260430-004Refund dispute — broken headphoneslow
- RX-260430-003SaaS landing page auditlow
- RX-260430-002AAPL & NVDA — weekly paper trade thesismedium
- RX-260430-001paper tradingmedium
- GLOBALpinnedused 2h ago[demo] Operator timezone: Europe/London (UK)Schedules and follow-ups should default to UK time unless the user specifies otherwise.
- CREWpinnedused 2h ago[demo] Always state simulation only on paper tradesEvery paper-trade output must clearly state it is simulation only. No 'execute', 'fill', or live-trading language. Risk under 1.5% of book per trade.
- CREWpinnedused 2h ago[demo] Prefer calm, firm tone for refund disputesWhen drafting refund or warranty disputes, default to a calm but firm voice with a 7-business-day window. Avoid ALL CAPS, exclamation marks, or threatening legal action.
- CONTACTcreated 2h ago[demo] Vendor 'Acme' takes 5–7 business days to respondRecurring vendor 'Acme' has a typical 5–7 business day response window for refund tickets. Set follow-up at +5d, escalate at +10d.
- CREWused 2h agoApproved paper-trade patternOperator approved a simulated trade with explicit invalidation and stop. Keep this pattern: thesis → invalidation → stop → target → confidence.
- CREWused 2h ago[demo] Bug Hunter — concise bullets in reportsQA reports should be brief: bullets, not prose. One line per issue, severity prefix, link to step.