The Rocky wiki for Hermes Agent.

The practical loop Rocky uses to turn requests into finished work: clarify scope, inspect sources, act with tools, verify output, and report what changed.

How to show terminal workflows publicly while hiding usernames, home folders, tokens, and machine-specific paths.

The once-only distribution pattern for videos, AI News posts, wiki pages, projects, and other HeyRocky-owned content.

The rules for turning external AI/dev articles into short HeyRocky summaries without republishing the original work.

How HeyRocky separates publishing, commenting, resharing, and AI News curation into small scheduled workers with safety rules.

A template for expanding HeyRocky wiki articles into useful public pages with summaries, screenshots, code blocks, and related links.

A practical guide to hermes agent overview in a verified Hermes/Rocky workflow.

A practical guide to install hermes on macos in a verified Hermes/Rocky workflow.

A practical guide to install hermes on linux in a verified Hermes/Rocky workflow.

A practical guide to install hermes on windows and wsl in a verified Hermes/Rocky workflow.

A practical guide to run the setup wizard in a verified Hermes/Rocky workflow.

A practical guide to choose a model provider in a verified Hermes/Rocky workflow.

A practical guide to openrouter provider setup in a verified Hermes/Rocky workflow.

A practical guide to anthropic provider setup in a verified Hermes/Rocky workflow.

A practical guide to openai codex provider setup in a verified Hermes/Rocky workflow.

A practical guide to gemini provider setup in a verified Hermes/Rocky workflow.

A practical guide to local model provider setup in a verified Hermes/Rocky workflow.

A practical guide to credential pools in a verified Hermes/Rocky workflow.

A practical guide to hermes cli command reference in a verified Hermes/Rocky workflow.

A practical guide to interactive chat sessions in a verified Hermes/Rocky workflow.

A practical guide to one-shot queries in a verified Hermes/Rocky workflow.

A practical guide to resume and continue sessions in a verified Hermes/Rocky workflow.

A practical guide to slash commands in a verified Hermes/Rocky workflow.

A practical guide to toolsets overview in a verified Hermes/Rocky workflow.

A practical guide to terminal tool in a verified Hermes/Rocky workflow.

How Rocky uses file tools safely to read, search, write, and patch project files without accidental broad edits.

A practical guide to web search and web extract in a verified Hermes/Rocky workflow.

A practical guide to browser automation in a verified Hermes/Rocky workflow.

A practical guide to macos computer use in a verified Hermes/Rocky workflow.

How Rocky uses vision tools to inspect screenshots, images, layouts, and visual evidence while still verifying important claims.

A practical guide to image generation tool in a verified Hermes/Rocky workflow.

A practical guide to text-to-speech tool in a verified Hermes/Rocky workflow.

A practical guide to speech-to-text in a verified Hermes/Rocky workflow.

A practical guide to elevenlabs voice narration in a verified Hermes/Rocky workflow.

A practical guide to discord gateway setup in a verified Hermes/Rocky workflow.

A practical guide to telegram gateway setup in a verified Hermes/Rocky workflow.

A practical guide to slack gateway setup in a verified Hermes/Rocky workflow.

A practical guide to webhook gateway in a verified Hermes/Rocky workflow.

A practical guide to gateway home channels in a verified Hermes/Rocky workflow.

A practical guide to cron delivery targets in a verified Hermes/Rocky workflow.

A practical guide to persistent memory in a verified Hermes/Rocky workflow.

A practical guide to user profile memory in a verified Hermes/Rocky workflow.

How structured fact storage supports deeper Rocky recall, entity reasoning, and memory hygiene beyond simple notes.

A practical guide to session search in a verified Hermes/Rocky workflow.

A practical guide to skills overview in a verified Hermes/Rocky workflow.

A practical guide to create a skill in a verified Hermes/Rocky workflow.

A practical guide to patch a skill in a verified Hermes/Rocky workflow.

A practical guide to skill curation in a verified Hermes/Rocky workflow.

How Hermes profiles separate projects, clients, credentials, memory, skills, and runtime settings for safer Rocky operations.

A practical guide to profile-safe paths in a verified Hermes/Rocky workflow.

How Hermes plugins extend Rocky with reusable integrations while keeping capabilities, dependencies, and operational boundaries clear.

A practical guide to native mcp client in a verified Hermes/Rocky workflow.

A practical guide to mcp server setup in a verified Hermes/Rocky workflow.

How Hermes cron jobs let Rocky run recurring briefs, checks, and automations safely with self-contained prompts and delivery rules.

A practical guide to script-only cron jobs in a verified Hermes/Rocky workflow.

A practical guide to attach cron jobs to sessions in a verified Hermes/Rocky workflow.

A practical guide to delegation and subagents in a verified Hermes/Rocky workflow.

A practical guide to background processes in a verified Hermes/Rocky workflow.

A practical guide to long-running agent spawns in a verified Hermes/Rocky workflow.

A practical guide to github operations in a verified Hermes/Rocky workflow.

A practical guide to google workspace in a verified Hermes/Rocky workflow.

A practical guide to discord collaboration patterns in a verified Hermes/Rocky workflow.

A practical guide to shopify operations boundaries in a verified Hermes/Rocky workflow.

A practical guide to randomshirts workflow pattern in a verified Hermes/Rocky workflow.

A practical guide to domain research workflow in a verified Hermes/Rocky workflow.

A practical guide to dynadot bridge pattern in a verified Hermes/Rocky workflow.

A practical guide to public website security sweep in a verified Hermes/Rocky workflow.

A practical guide to private portal security sweep in a verified Hermes/Rocky workflow.

A practical guide to secret redaction in a verified Hermes/Rocky workflow.

A practical guide to pii redaction in a verified Hermes/Rocky workflow.

A practical guide to approval modes in a verified Hermes/Rocky workflow.

A practical guide to safe handling of passwords in a verified Hermes/Rocky workflow.

How audit logs make Rocky workflows reviewable by recording important actions, approvals, sources, and verification evidence.

A practical guide to csp and security headers in a verified Hermes/Rocky workflow.

A practical guide to robots and indexing in a verified Hermes/Rocky workflow.

A practical guide to seo content workflow in a verified Hermes/Rocky workflow.

A practical guide to public vs private pages in a verified Hermes/Rocky workflow.

A practical guide to knowledge base article template in a verified Hermes/Rocky workflow.

A practical guide to video script template in a verified Hermes/Rocky workflow.

A practical guide to screen recording checklist in a verified Hermes/Rocky workflow.

A practical guide to narration audio workflow in a verified Hermes/Rocky workflow.

A practical guide to video editing checklist in a verified Hermes/Rocky workflow.

A practical guide to youtube upload checklist in a verified Hermes/Rocky workflow.

A practical guide to document production in a verified Hermes/Rocky workflow.

A practical guide to pdf verification in a verified Hermes/Rocky workflow.

A practical guide to powerpoint and slides in a verified Hermes/Rocky workflow.

A practical guide to website qa checklist in a verified Hermes/Rocky workflow.

A practical guide to mobile responsive qa in a verified Hermes/Rocky workflow.

A practical guide to console and network qa in a verified Hermes/Rocky workflow.

A practical guide to deployment verification in a verified Hermes/Rocky workflow.

A practical guide to cloud run hosting in a verified Hermes/Rocky workflow.

A practical guide to dns and domain safety in a verified Hermes/Rocky workflow.

A practical guide to backups and recovery in a verified Hermes/Rocky workflow.

A practical guide to safe lockdown pattern in a verified Hermes/Rocky workflow.

A practical guide to hermes doctor troubleshooting in a verified Hermes/Rocky workflow.

A practical guide to gateway troubleshooting in a verified Hermes/Rocky workflow.

A practical guide to provider troubleshooting in a verified Hermes/Rocky workflow.

A practical guide to tool not available troubleshooting in a verified Hermes/Rocky workflow.

A practical guide to model fallback testing in a verified Hermes/Rocky workflow.

A practical guide to usage analytics in a verified Hermes/Rocky workflow.

A practical guide to cost and quota monitoring in a verified Hermes/Rocky workflow.

A practical guide to public knowledge base roadmap in a verified Hermes/Rocky workflow.

A practical guide to jarvis-style system architecture in a verified Hermes/Rocky workflow.

A practical guide to private ai for law firms in a verified Hermes/Rocky workflow.

A practical guide to restaurant command center in a verified Hermes/Rocky workflow.

A practical guide to personal operations memory in a verified Hermes/Rocky workflow.

A practical guide to ai-assisted security reviews in a verified Hermes/Rocky workflow.

A practical guide to ai-assisted website builds in a verified Hermes/Rocky workflow.

Use this checklist before trying to install Rocky into a real business workflow.

Design approval queues so Rocky can move fast without doing risky things silently.

What belongs in Rocky memory, what belongs in a file, and what should never be stored.

How to decide which Rocky tools are safe for a workflow and where human approval is required before taking action.

Use Discord as a command and delivery surface without turning it into a noisy dumping ground.

How to create daily Rocky briefings that are worth reading by focusing on changes, urgency, approvals, and next actions.

How Rocky should research public web sources with source links, uncertainty notes, freshness checks, and public-safe summaries.

How Rocky can use the Mac in the background safely while preserving user control, privacy, and approval boundaries.

A practical checklist for verifying public HeyRocky pages before launch: layout, links, metadata, images, console, and mobile behavior.

Turn scattered customer requests into a clean Rocky-ready intake flow.

Use Rocky to draft customer replies without sending risky messages automatically.

How to keep Rocky run logs useful later with clear requests, sources, actions, verification, and privacy-safe evidence.

How to help visitors find the right guide without scrolling a giant list.

How to turn repeated Rocky/Hermes work into reusable skills with triggers, steps, pitfalls, and verification checks.

Safely handling API keys, tokens, and provider credentials in Rocky systems.

A practical pipeline for images, videos, and tutorials that do not ship obvious mistakes.

How to design recurring Rocky jobs that alert only when useful, stay quiet when nothing changed, and avoid duplicate notifications.

How to keep HeyRocky public pages and private dashboards separated so visitors see only the right content and actions.

What a private Rocky dashboard should show first: approvals, exceptions, recent changes, and decisions that need owner attention.

How to diagnose model provider setup problems in Rocky and Hermes, including credentials, model names, quotas, and live verification.

How to convert a strong wiki page into a tutorial video plan with a clean script, visuals, narration, and QA checks.

Keep Rocky useful without burning tokens, credits, or image/video budget unnecessarily.

A practical architecture for using Rocky and Hermes as an AI operator system: intake, context, tools, approvals, verification, and reporting.

How to decide which Rocky and Hermes actions can run automatically and which should stop for explicit human approval.

A finish-the-job checklist for Rocky workflows: inspect the source, act with the right tool, verify with evidence, and report clearly.

How to list the files, dashboards, APIs, docs, accounts, and people Rocky should treat as authoritative before acting.

A practical triage pattern for converting vague user requests into scoped Rocky work orders with constraints, tools, and verification steps.

Clear boundaries for unsafe, unsupported, ambiguous, private, financial, legal, or destructive requests in Rocky workflows.

How to design a daily Rocky brief that summarizes what changed, what is urgent, what needs approval, and what can wait.

A lightweight logging pattern for Rocky runs: request, sources inspected, actions taken, proof, approvals, and blockers.

A design guide for private Rocky command centers that show status, drafts, approvals, alerts, and recent verified actions.

A capability ladder that explains the difference between simple chat, assisted workflows, tool use, verified operations, and supervised automation.

A practical guide to the Hermes configuration file: model, tools, gateway, memory, security, terminal, compression, and display settings.

How to use Hermes profiles to separate personal work, client work, experiments, production workflows, tools, skills, and memory.

How to think about provider authentication, credential pools, quota exhaustion, and fallback models in Hermes without exposing secrets.

A practical model-selection guide for Rocky workflows: coding, web research, writing, vision, media, automation, and low-cost summaries.

A beginner-friendly guide to enabling Hermes tools gradually: web, file, terminal, browser, vision, image generation, gateway, and automation tools.

How to document and use environment variables safely in Rocky and Hermes workflows without exposing keys, tokens, or private paths.

The post-install checklist for confirming Hermes and Rocky are ready: setup wizard, model auth, doctor checks, tools, and a harmless test task.

A safe update procedure for Hermes users with production workflows, scheduled jobs, gateway bots, skills, and custom configurations.

How to understand where Hermes terminal and file operations are running, especially when using local, Docker, SSH, or other remote backends.

How to use Hermes from the CLI, phone, Discord, Telegram, Slack, and other gateway surfaces while keeping routing and approvals clear.

What Rocky should remember, what it should avoid remembering, and how to keep long-running assistant memory useful and safe.

How to recognize when a Rocky workflow should become a reusable Hermes skill with triggers, steps, pitfalls, and verification.

A lifecycle guide for keeping Hermes skills accurate: create after useful discoveries, patch after failures, and retire stale procedures safely.

A practical pre-flight checklist for turning a Rocky idea into a bounded automation with clear inputs, safe actions, rollback, and proof of completion.

How to decide between durable scheduled Hermes jobs, one-off background processes, and normal foreground tool execution for Rocky operations.

A publishing checklist for Rocky wiki pages that are useful, grounded, indexable, and safe for public readers.

How Rocky decides when it can act autonomously and when a human approval boundary is the safer operating mode.

A practical guide to choosing primary and fallback model providers without hard-coding secrets or building workflows around one fragile endpoint.

How to decide which tools Rocky should use for a workflow, from read-only research to file edits, desktop automation, image generation, and deployment.

A lightweight incident note pattern for AI-assisted operations: what failed, what evidence proved it, how it was fixed, and how to avoid a repeat.

A visual QA checklist for catching the issues automated tests miss: awkward wrapping, tiny chips, clipped text, duplicate images, and unprofessional layout details.

How Hermes profiles help keep different Rocky work contexts, tools, memories, and credentials separated instead of blending every task together.

A practical way to keep a growing Rocky wiki searchable as article count rises: query sets, result relevance, snippets, filters, and sitemap coverage.

A public-safe checklist for connecting MCP servers to Rocky through Hermes without guessing at tool availability or exposing credentials.

How to design public-safe event-driven Rocky workflows with clear payload boundaries, test events, and delivery verification.

A guide to using Hermes session history as evidence of what was discussed, while still checking live sources before making current claims.

When to use quick delegated subagents, background terminal jobs, or full spawned Hermes agents for longer Rocky workstreams.

A practical QA pattern for combining browser automation, screenshots, and human-eye checks on public Rocky pages.

How Rocky should handle image generation requests when the approved provider is unavailable, especially for public wiki illustrations.

A reporting pattern for scheduled Rocky jobs: when to stay silent, when to send a concise report, and what evidence belongs in the final output.

How to ship public wiki/search data with a Cloud Run service without accidentally including private datastore files.

A recovery-minded approach to risky Rocky work: checkpoints, small changes, visible diffs, and rollback-ready verification.

A public-safe troubleshooting guide for checking Rocky/Hermes health before guessing at model, tool, gateway, or deployment failures.

A checklist for making new Rocky wiki pages discoverable without creating duplicate, stale, or incorrect indexable URLs.

Choose the right Hermes tool for file work, browsing, desktop control, media generation, and verification without crossing safety boundaries.

A practical guide to choosing model providers, fallback paths, and cost tiers for Rocky without making unsupported claims.

How to add Model Context Protocol servers to Rocky with scoped tools, tests, and rollback notes.

Design webhook-triggered Rocky workflows with authentication, payload limits, replay protection, and human-readable delivery reports.

A public-safe runbook for recurring Rocky jobs that gather facts, summarize changes, and report only when there is something useful.

Use background desktop automation for screenshots and clicks while respecting permissions, passwords, and user focus.

A release checklist for keeping public HeyRocky pages separate from private workspaces, customer data, and internal automation state.

A Cloud Run release checklist for public wiki updates: build, deploy, route probes, sitemap checks, and visual smoke testing.

Turn repeated Rocky workflows into reusable skills with triggers, commands, pitfalls, and verification rather than stale session notes.

Prepare Rocky for Telegram, Discord, Slack, voice, and other gateway channels with clear status checks and fallback plans.

A practical gate checklist for getting Rocky ready to execute real work safely: model routing, tools, memory, approvals, channels, and verification.

How to grow a public Rocky wiki with strong pages, privacy review, search checks, screenshots, sitemap coverage, and live verification.

A practical guide to choosing approval modes and command boundaries when Rocky operates files, terminals, browsers, and deployments.

Choose Hermes toolsets intentionally so Rocky has enough capability to act without exposing unnecessary surfaces.

When to use user memory, procedural skills, project files, and official docs so Rocky avoids stale or private context mistakes.

Plan Model Context Protocol server adoption with safe tool discovery, scoped credentials, testing, and rollback.

Prepare webhook-triggered Rocky workflows with signing, idempotency, delivery routing, logging, and public/private boundaries.

Use Hermes profiles to separate client work, worker agents, credentials, skills, memories, and delivery surfaces.

Choose model tiers, route premium work intentionally, and handle provider failures without unsupported promises.

Create a compact troubleshooting packet with reproduction steps, source checks, logs, screenshots, and rollback notes before changing production.

Run scheduled Rocky work like a release process: inspect sources, edit only safe surfaces, verify locally, deploy only on green checks, and report receipts.

Make Rocky reports trustworthy by including what was checked, what changed, what returned, and what still needs human attention.

Move public knowledge base content from grounded source material to indexed pages with canonical URLs, search inclusion, and sitemap coverage.

Turn repeatable small-business work into Rocky runbooks with inputs, tools, approvals, verification, and escalation rules.

Design approval gates so Rocky can move quickly on low-risk work while stopping for payments, secrets, destructive commands, and sensitive messages.

Use browser automation to verify public pages like a visitor: route status, console errors, mobile layout, images, forms, and visual contact sheets.

Keep durable memory useful by saving stable preferences and environment facts while avoiding stale task logs, private details, and temporary progress notes.

Generate knowledge-base visuals that look premium without exposing private screens, fake UI labels, unreadable text artifacts, or off-brand mascots.

Use subagents and worker processes safely by giving each worker a narrow task, enough context, and clear limits on side effects.

Use terminal access responsibly: prefer read-first discovery, avoid destructive commands, keep secrets out of logs, and verify command results.

Choose and operate gateway, plugin, and shell hooks without turning lifecycle automation into an invisible or over-privileged failure point.

Design event-driven Rocky workflows with signed routes, narrow prompts, idempotency, rate limits, and least-privilege tool access.

Use Hermes no-agent cron mode for deterministic watchdogs, alerts, and metrics that should run cheaply and stay silent when healthy.

Prevent surprise cost and authentication failures by treating scheduled-job model selection, credentials, workdirs, and prompts as release-time configuration.

Run builds, tests, servers, and interactive CLIs without losing process visibility, completion signals, logs, or cleanup responsibility.

Let Rocky operate desktop apps without taking over the user's cursor while preserving approvals, fresh accessibility captures, and visual verification.

Operate the Hermes dashboard as a localhost management surface, keep profile write targets explicit, and harden any deliberate remote access.

Install and configure MCP servers with reviewed sources, narrow tool filters, explicit transport trust, and real authenticated tool-call verification.

Expose Hermes conversation and approval workflows to another MCP-capable client while preserving gateway dependencies, event cursors, and send boundaries.

Package a reusable Hermes agent as a versioned Git repository while keeping installer secrets, memories, sessions, runtime databases, and local overrides private.

Use Hermes Kanban when work must survive restarts, cross profile boundaries, preserve handoffs, and remain visible to human reviewers.

Design named Hermes worker lanes with explicit assignees, pinned workspaces, lifecycle termination, durable evidence, and review-required handoffs.

Use persistent goals for multi-turn missions, then tighten autonomous progress with verification criteria, scope boundaries, subgoals, and wait barriers.

Use project and subdirectory context files to give Hermes stable architecture, conventions, commands, and safety boundaries without repeating them every turn.

Attach exact files, line ranges, folders, diffs, commits, or public URLs to a CLI prompt while respecting workspace, size, and sensitive-path limits.

Return verified charts, documents, spreadsheets, images, audio, video, and archives as native messaging attachments instead of unusable local paths.

Run isolated prompt datasets through Hermes for trajectory generation or evaluation with explicit concurrency, checkpoints, tool distributions, and cost controls.

Reduce repeated schema overhead by loading MCP and plugin tools on demand while preserving real approvals, hooks, guardrails, and tool identities.

Expose a tool-equipped Hermes profile to approved local clients through an authenticated OpenAI-compatible API with deliberate session and network boundaries.

Run Hermes as an Agent Client Protocol server for editor-native coding with curated tools, stdio transport, visible approvals, and provider reuse.

Catch type errors, undefined names, missing imports, and project-wide semantic issues after file edits by pairing syntax checks with language servers.

Keep several Hermes agents online with isolated credentials, sessions, memory, and services, or deliberately choose a single multiplexing gateway.

Pin a small administrator-owned baseline of Hermes configuration and secrets without freezing every user-controlled setting.

Give each coding agent an isolated branch and checkout so parallel work cannot overwrite another agent’s files or rollback history.

Opt into shadow-store snapshots before file mutations, preview changes, restore one file or a full checkpoint, and keep storage bounded.

Operate Hermes in the official container with persistent state, supervised gateways, non-root execution, bounded network exposure, and real health checks.

Load provider credentials from Bitwarden, 1Password, or vetted plugins at startup while preserving provenance and deterministic conflict rules.

Scan the active environment, plugin requirements, and pinned MCP packages for known vulnerabilities while keeping audit results actionable.

Reproduce agent failures without user config, project rules, memory, skills, plugins, hooks, or MCP servers before changing a healthy installation.

Use offline prompt-size reporting to find oversized skill indexes, memory, profile guidance, and tool schemas before context pressure becomes a reliability problem.

Read the correct agent, gateway, desktop, error, or update log for the same profile and time window as the failure.

Use the modern terminal interface for live sessions, model and theme controls, subagent observability, approvals, and shared session history.

Use Hermes Desktop as a native interface to the same profiles, files, sessions, tools, voice, and memory already configured for CLI and gateway use.

Let local OpenAI-compatible applications use a Hermes-managed OAuth subscription for raw model inference without exposing the full agent toolset.

Choose a full installation backup or credential-stripped profile export, restore to a new machine, and verify providers, paths, sessions, and gateways.

Map the entry points, agent core, provider runtime, tool registry, session store, and gateway before changing or troubleshooting Hermes.

Follow a Hermes turn from prompt assembly through API-mode conversion, tool execution, callbacks, persistence, interruption, and fallback.

Preserve long-session quality by separating context compression, gateway hygiene, protected message tails, summary routing, and provider prompt caching.

Trace authorization, session keys, interruption, slash commands, adapters, delivery, hooks, and profile-scoped gateway processes.

Operate Hermes session persistence safely across SQLite storage, full-text search, profile boundaries, compression lineage, exports, and cleanup.

Understand how Hermes resolves provider, model, API mode, base URL, credential source, auxiliary routes, and fallback without leaking keys across endpoints.

Decide when core code is justified, then register a schema, availability check, handler, toolset membership, tests, and safe error contract.

Choose a custom, plugin, or native path and verify auth, model discovery, runtime resolution, auxiliary routing, protocol adapters, tests, and docs.

Use the plugin-first adapter path to connect a messaging service with authorization, scoped locks, delivery, cron support, platform hints, and parity tests.

Build wrapper CLI widgets, keybindings, commands, layouts, and styles through protected extension hooks instead of replacing the main run loop.

Control skill-library growth with dry runs, inactivity states, pins, backups, rollback, optional consolidation, and auditable curator reports.

Choose one additive external memory system by storage, privacy, recall, extraction, cost, identity, and profile-isolation requirements.

Understand stable, contextual, volatile, and API-time prompt layers so customization remains predictable and cache-friendly.

Build a maintainable Hermes plugin with a clear manifest, precise tool schemas, safe handlers, lifecycle hooks, and a verification path.

Implement a cloud-browser provider that owns remote session lifecycle while Hermes keeps the common browser tool surface.

Replace the default context compressor with an explicit, testable engine while preserving valid message history and session lifecycle behavior.

Add an image backend behind the shared `image_generate` tool with capability-aware routing, normalized responses, and setup-ready model metadata.

Implement a video backend that routes text-to-video and image-to-video through one capability-aware Hermes tool contract.

Connect a search or extraction backend to Hermes with explicit capability flags, normalized results, and independent search/extract routing.

Use the host-owned `ctx.llm` surface for bounded plugin-side model calls without duplicating provider credentials or bypassing operator trust controls.

Follow a Hermes tool from self-registration and toolset filtering through availability checks, central dispatch, hooks, and structured error handling.

Read, validate, and handle Hermes ShareGPT-compatible trajectory files for debugging, evaluation, and training-data workflows.

Choose between ACP, the TUI gateway, and the OpenAI-compatible API server when embedding Hermes in an IDE, custom host, or web client.

Connect Rocky to Meta’s official WhatsApp Business Cloud API with signed webhooks, explicit sender access, stable credentials, and production-aware delivery checks.

Operate Rocky through Signal using a linked signal-cli daemon, deny-by-default access, scoped group controls, attachment checks, and reconnect verification.

Give Rocky a dedicated email inbox with IMAP polling, SMTP replies, strict sender controls, attachment policy, threading tests, and duplicate-reply prevention.

Deploy Rocky on Matrix with room and user allowlists, mention policy, thread-aware sessions, optional encryption, and bridge loop protection.

Prepare a Rocky bot for Microsoft Teams with a public HTTPS endpoint, tenant-scoped credentials, AAD user controls, health checks, and stable production routing.

Connect Rocky to Home Assistant with focused event filters, explicit device actions, blocked high-risk service domains, and reversible automation tests.

Use Open WebUI as a polished client for Hermes while keeping API authentication, execution-host boundaries, model discovery, and profile isolation explicit.

Use ntfy as a lightweight Rocky push channel with topic-based identity, access-controlled publishing, one-way alert mode, and cron delivery verification.

Move from installation to one verified Hermes conversation before adding gateways, automation, voice, or advanced routing.

Choose the right Hermes documentation sequence for a CLI assistant, messaging bot, automation operator, extension author, or embedded application.

Install and verify the tested Tier 2 Hermes path on Android while respecting Termux-specific dependency and voice limitations.

Choose between Nix profile installation, a native NixOS service, and a persistent container while keeping state and secrets explicit.

Use one OAuth-backed Nous Portal setup for model access and managed web, image, speech, browser, and optional sandbox tools.

Connect Hermes to llama.cpp or an MLX server on Apple Silicon with realistic memory, context, networking, and verification gates.

Use AIAgent inside Python services and pipelines with explicit tool allowlists, isolated instances, quiet output, and bounded authority.

Use SOUL.md for stable identity and voice while keeping project rules, permissions, and temporary modes in their proper layers.

Configure speech input and spoken replies across CLI, Telegram, Discord text, and Discord voice channels with provider and audio-path verification.

Group compatible skills behind one short command while preserving load order, trust review, and a clear verification contract.

Evaluate and enable bundled Hermes plugins explicitly, with attention to discovery order, hooks, authority, observability, and rollback.

Use Hermes project workspaces to group related folders and align desktop sessions, kanban tasks, branches, and worktree conventions.

Choose how follow-up messages affect an active Hermes run so urgent corrections, patient follow-ups, and in-flight guidance behave predictably.

Run the Hermes web dashboard as an administrative control plane with explicit bind, authentication, profile, secret, and restart boundaries.

Use the offline prompt-size report to find oversized skills, memory, profile context, or tool schemas before they crowd out working context.

Collect reproducible Hermes diagnostics with dump, debug, logs, and redaction while keeping secrets and unrelated private content out of support artifacts.

Authorize messaging users with expiring pairing codes, explicit approval, revocation, and platform-specific unauthorized-DM behavior.

Preserve a useful conversation while exploring alternatives, moving between surfaces, and returning to the newest session lineage without losing provenance.

Choose quick state snapshots or full Hermes-home backups, verify archive integrity, and rehearse restoration before upgrades or migrations.

Inspect, pause, resume, and recover individual Hermes messaging adapters while preserving healthy channels and understanding restart-interrupted sessions.

Keep session search useful by exporting, archiving, pruning, and verifying conversation history with explicit age, platform, lineage, and privacy rules.

Recognize context-pressure warnings, reduce avoidable prompt and tool-output growth, and compress or branch work without losing the completion contract.

Run an authorization-first iMessage bridge through BlueBubbles while keeping credentials, group behavior, and delivery checks explicit.

Connect Hermes to Google Chat through a least-privilege GCP service account and Pub/Sub path, then prove identity and delivery behavior end to end.

Configure a LINE Messaging API channel with a stable HTTPS webhook, strict sender policy, and verified media-delivery behavior.

Create a dedicated Mattermost bot account, constrain its channel membership, and verify user and thread isolation before team rollout.

Connect Hermes to IRC with deliberate nickname, TLS, channel, mention, and access-control policies suited to a public-by-default protocol.

Operate a SimpleX Chat gateway through a local daemon with pairing, group controls, attachment checks, and cron delivery boundaries.

Create a Feishu or Lark app, request only needed permissions, and operate the Hermes adapter through a resilient long connection.

Connect a WeCom AI Bot to Hermes with QR-assisted setup, explicit access policy, and realistic streaming and delivery expectations.

Use the Weixin QR login flow without exposing account state, then verify authorization, media, multiline, and restart behavior.

Operate Hermes over Twilio SMS with exact webhook matching, request-signature validation, E.164 allowlists, and delivery-status checks.

Receive Microsoft Graph notifications through Hermes with clientState validation, HTTPS termination, minimal responses, and subscription lifecycle checks.

Build a Microsoft Teams meeting pipeline that prefers transcripts, controls recording fallback, preserves durable state, and verifies every delivery sink.

Resolve provider credentials from reviewed 1Password references at startup, with deliberate authentication, cache, rotation, and non-interactive runtime checks.

Use a least-privilege Bitwarden machine account to hydrate Hermes provider credentials at startup and rotate them centrally without broad vault access.

Bridge an existing CLI vault or protected runtime file into Hermes through a fast non-interactive helper that emits bounded KEY=VALUE output.

Compose local, mapped, bulk, and profile-aliased secret sources with deterministic ownership so one credential never wins by accident.

Rotate provider keys and secret-manager bootstrap tokens as a verified runtime change, with rollback, restart, provenance, and failure-mode checks.

Choose cache TTLs and network-failure behavior for external secret sources without allowing revoked credentials or partial fetches to masquerade as healthy state.

Finish loopback OAuth for remote MCP and service integrations through paste-back or an SSH local forward while preserving exact callback state and port boundaries.

Operate system-browser sign-in for gated Hermes desktops with PKCE, loopback callbacks, rotating tokens, OS-keychain storage, and explicit fallback checks.

Integrate an external vault as a standalone Hermes plugin while leaving precedence, timeouts, provenance, protected variables, and environment writes to the core orchestrator.

Verify which source supplied a provider credential, how Hermes represented it in the credential pool, and whether fallback or rotation changed the active identity.

Update Hermes deliberately with preview, state snapshots, branch awareness, gateway restart checks, and a tested rollback path.

Select macOS, native Windows, WSL2, Linux, Docker, Android, or Nix with an explicit support-tier and feature-boundary decision.

Operate Hermes inside WSL2 without slow repositories, broken permissions, line-ending surprises, or confused Windows-to-Linux networking.

Install and validate Hermes directly on Windows while respecting Git Bash execution, locked-process updates, UTF-8, and the embedded-terminal limitation.

Move compatible OpenClaw state into Hermes through a previewed, backed-up migration while keeping secrets, conflicts, and re-pairing explicit.

Set up a profile-safe Hermes development environment, choose the right extension surface, test cross-platform behavior, and prepare a reviewable contribution.

Develop TUI and desktop branches from isolated worktrees while safely sharing dependencies only when lockfiles match.

Customize Hermes CLI colors, spinner language, branding, and tool accents without confusing visual identity with agent personality.

Add an optional profile-scoped animated pet to Hermes, choose a rendering mode, and verify activity states without affecting agent behavior.

Trace how Hermes adapts synchronous agent runs into ACP editor sessions with workspace binding, streamed events, cancellation, and fail-closed approvals.

Configure Hermes for Amazon Bedrock’s native Converse API with IAM credentials, regional model discovery, inference profiles, and a verified first request.

Connect Hermes to Microsoft Foundry or Azure OpenAI with API-key or Microsoft Entra ID authentication and the correct endpoint transport.

Route Hermes to Gemini on Vertex AI with service-account or Application Default Credentials and automatically refreshed OAuth tokens.

Use Hermes’ xAI device-code OAuth path for an eligible SuperGrok or X Premium+ account, with explicit entitlement and API-key fallback checks.

Choose, configure, and verify Hermes’ opt-in Codex app-server runtime while preserving clear tool, sandbox, plugin, and rollback boundaries.

Connect Hermes to Ollama’s local OpenAI-compatible endpoint with realistic hardware, context-window, tool-calling, privacy, and fallback checks.

Connect the global MiniMax provider through Hermes’ browser or headless OAuth flow and verify token refresh, region, and model selection.

Build a named Mixture of Agents preset with advisory reference models, one acting aggregator, and explicit cost, cadence, and privacy controls.

Keep real provider keys on the host while Docker sandboxes receive opaque proxy tokens through Hermes’ fail-closed iron-proxy integration.

Design a narrow outbound-host policy, safe credential source, and auditable deny posture for Hermes Docker sandboxes.

Diagnose install, listener, CA, mapping, and provider failures while preserving the egress proxy’s fail-closed Docker boundary.

Create a maintainable dashboard theme with deliberate palette, typography, layout, assets, and rollback checks instead of fragile global CSS.

Add dashboard tabs, shell slots, or page augmentations with a manifest, SDK frontend, scoped backend routes, and explicit reload verification.

See how persistent CDP supervision surfaces blocking JavaScript dialogs and cross-origin iframe targets without bloating unsupported browser sessions.

Launch and supervise plugin-owned child sessions through opaque handles, narrow toolsets, cooperative cancellation, and honest process-lifetime limits.

Trace job storage, tick locking, fresh-session execution, provider triggers, delivery, recursion guards, and scale-to-zero scheduling.

Choose stdio or HTTP MCP transport, preserve certificate verification, configure mTLS or OAuth, and fail fast on unsafe connection settings.

Filter MCP tools, resources, and prompts deliberately; understand include precedence, dynamic toolsets, naming, and empty-registration behavior.

Use priority processing only when faster model responses justify provider-specific cost and availability tradeoffs, then verify the active mode before important work.

Move an active CLI conversation to a configured messaging home without losing its role-aware transcript, then verify the new thread before leaving the terminal.

Use Telegram DM topics as separate user-managed Hermes workspaces, with explicit session restoration and safe controls for turning topic mode off.

Show model, context, and working-directory metadata when it helps operators, and hide it when clean external-facing replies matter more.

Use /learn to distill repeatable knowledge from a directory, URL, completed workflow, or notes—then inspect and test the resulting skill before reuse.

Generate or merge-update a concise AGENTS.md from a repository scan so future agents inherit real project commands, structure, and boundaries.

Evaluate suggested automations as proposed cron jobs, preserve the right delivery origin, and accept only bounded work with explicit safety and verification rules.

Use blueprint templates to create repeatable cron jobs through guided or direct slot filling, then inspect the concrete job before its first scheduled run.

Let goal loops wait on tracked builds and processes instead of wasting turns, with durable barriers, completion notifications, and stale-wait recovery.

Keep session history useful with deliberate naming, search, resume, redacted export, and retention—without treating old transcripts as current system truth.

Use execute_code for multi-tool Python workflows while preserving tool limits, environment isolation, and concise final output.

Configure Honcho memory, choose context and dialectic cadence, isolate peers, and verify the right context reaches each session.

Control aggregator sub-provider selection by price, latency, throughput, capability support, and data-collection policy.

Preview and migrate instructions, permission rules, MCP servers, skills, and memories from Claude Code or Codex CLI without importing credentials.

Diagnose cron timing, gateway, delivery, skill, timeout, permission, and lock problems with a repeatable fault-isolation sequence.

Use hermes send to deliver raw script, file, CI, and monitoring output through configured messaging adapters without an LLM.

Trigger sandboxed pull-request reviews from signed GitHub events, fetch the real diff, and post concise idempotent comments.

Enable an on-device wake phrase, choose the listening surface and engine, and tune false-trigger and microphone behavior.

Enable Spotify tools, complete PKCE OAuth, understand Free versus Premium actions, and verify playback on an active Connect device.

Use the read-only x_search tool for cited X discovery, validate filters, and recognize degraded unsourced responses.

Pair the unofficial WhatsApp bridge, isolate account risk, persist sessions safely, and enforce sender access controls.

Choose among Buzz Desktop, the ACP relay bridge, and Hermes’s native Buzz gateway while keeping identities and transports isolated.

Run Hermes in DingTalk DMs and groups over an outbound WebSocket, with per-user session isolation, mention gates, and optional AI Cards.

Use Photon’s managed iMessage lines and supervised Node sidecar without hosting a Mac relay or exposing a public webhook.

Connect through QQ Bot API v2 for C2C, groups, guilds, attachments, and voice transcription with explicit DM and group policies.

Use content-free wake hints to activate Hermes while the Raft CLI retains ownership of message bodies, cursors, and replies.

Front multiple messaging platforms over one outbound WebSocket while a separate connector retains platform credentials and advertises supported operations.

Operate Hermes as a first-class WeCom app using signed encrypted callbacks, proactive replies, public HTTPS, and optional multi-corporation routing.

Connect Hermes to Yuanbao over an HMAC-authenticated WebSocket with direct and group routing, media delivery, and a deliberate home channel.

Create a single-tenant Entra application for app-only Teams transcript access, grant least-privilege permissions, scope users, and rotate secrets safely.

Install Hermes under a dedicated non-root account by separating one-time browser system dependencies, the per-user installer, launcher PATH, and service-runtime verification.

Route Hermes terminal and file work into a Vercel Sandbox microVM with the supported SDK, authentication, runtime, snapshot, disk, and background-process settings.

Give a Hermes skill or sandbox the minimum environment variables and credential files it needs without inheriting the host's complete secret-bearing environment.

Select among Hermes' eight web backends, split search from extraction when useful, control auto-detection, and verify the active provider without configuration surprises.

Compare Chrome, Chromium, CDP, and browser-provider options by runtime, isolation, login continuity, and verification needs.

Resume the right Hermes session, confirm its identity, inspect workspace changes, measure what fills the context window, and reduce on-screen noise without discarding history.

Mine repeated approved commands into reviewable Hermes allowlist proposals, apply only selected patterns, and preserve deny rules and the hardline safety floor.

Use live agent activity, delegation transcripts, stall metadata, timeout diagnostics, and ownership rules to distinguish slow Hermes subagents from genuinely wedged work.

Recover failed or blocked Hermes Kanban work by reading run history, choosing a bounded retry budget, fixing the underlying fault, and verifying the next dispatch.

Set speech-recognition language, model, and fallback behavior for accurate Hermes voice input across local and remote environments.

Wire trusted local speech CLIs into Hermes with the official TTS and STT command-provider registries, correct placeholders, bounded timeouts, and explicit output verification.

Repair macOS Accessibility and Screen Recording permissions when Hermes Desktop or Computer Use cannot capture, click, or type reliably.

Keep routine Hermes CLI sessions readable, then switch to verbose output only when a provider, tool, or configuration problem needs deeper evidence.

Load known skills before the first turn so a Hermes session begins with the right workflow, commands, pitfalls, and verification rules already in context.

Run quick shell checks with the CLI exclamation-prefix path when no model reasoning is needed, while preserving normal approval and working-directory boundaries.

Bound a single Hermes turn, understand the final grace call, and tune transient provider retries so fallback happens at the intended speed.

Make Hermes continue after code edits when a turn lacks fresh test, build, lint, or equivalent evidence, with bounded nudges and surface-aware behavior.

Detect identical failures, repeated same-tool errors, and no-progress calls; warn interactive agents and hard-stop unattended runs before they waste a turn budget.

Preserve working directory and exported variables across terminal calls when the backend benefits, while understanding one-shot fallbacks and precedence.

Set provider-wide and model-specific timeouts without confusing total request limits, stale-call detection, stream gaps, or local-model prefill.

Protect the context window from giant files, minified lines, and noisy commands by tuning read, line, and byte caps while preserving paginated access.

Keep a durable objective active across responses while capping judge-driven continuation turns and preserving pause, resume, and human steering.

Reduce on-screen tool chatter in CLI and TUI sessions while preserving conversation history, tool results, prompt caching, and a visible recovery path.

Create reviewed no-model shortcuts for repeated shell actions across CLI and messaging surfaces, with narrow arguments, safe output, and normal approval boundaries.

Understand how the Hermes container chooses between the full s6 supervision tree and a direct stage-two fallback when another platform owns PID 1.

Separate Kanban dispatch ownership from notification delivery so task updates return through the gateway profile that owns the originating adapter.

Use Hermes confirmation controls to prevent accidental session clearing, reset, undo, or deletion while preserving an explicit one-time bypass for deliberate automation.

Park unfinished prompts with Ctrl+S, restore them later, and browse a session-only draft stack without losing multiline text or attached images.

Launch independent background prompt sessions while keeping the foreground interactive, with clear boundaries around inherited settings, history, and task results.

Draft long or formatting-sensitive prompts in your configured editor, then return them to Hermes without fighting a single-line terminal composer.

Set an installed terminal font or CSS font stack per Hermes profile while retaining the bundled JetBrains Mono fallback and avoiding remote font dependencies.

Give each Hermes profile a durable capability description so Kanban orchestration can route work by evidence instead of guessing from a profile name.

Create memorable profile wrapper commands and install shell completion without hiding which Hermes profile a command will use.

Create a minimal skill footprint or stop future bundled-skill seeding without deleting user-edited, hub-installed, or handwritten skills.

Build a hot-reloading Hermes Desktop extension with the official SDK while keeping disk-plugin imports and optional Python backends inside their documented security boundaries.

Implement the Hermes MemoryProvider contract with non-blocking synchronization, explicit privacy disclosure, profile-safe storage, and lifecycle verification.

Design a cross-provider failover chain that preserves sessions, controls auxiliary routing, and avoids costly fallback surprises.

Use Hermes model-catalog manifests, caching, and bundled snapshots to balance fresh curation with reliable offline selection.

Select core, composite, platform, and dynamic toolsets with a least-privilege workflow and clear verification gates.

Evaluate, install, verify, and retain optional Hermes skills one capability at a time instead of activating a broad catalog blindly.

Choose native configuration, plugins, or MCP by matching each integration to the boundary it actually needs.

A practical operating discipline for better Hermes prompts, cleaner persistent context, predictable token use, and safer tool execution.

A practical pattern for parallel Hermes subagents that preserves context isolation, prevents file collisions, and requires parent-level verification.

An architecture for scheduled GitHub pull-request analysis that begins report-only and adds authenticated review actions only behind explicit approval policy.

A staged MCP adoption plan that starts with one bounded server, allowlists only necessary tools, and verifies the visible capability surface after every change.

A safe method for refining Hermes identity while keeping project instructions, secrets, and profile-specific state in their proper boundaries.