Hermes Action Bridge

Hermes Action Bridge

A configurable bridge that lets external agents delegate real-world actions to Hermes Agent without reimplementing Hermes skills, tools, platform integrations, browser automation, cron jobs, or messaging flows.

Use it from Claude Code, Codex, Cursor, CI jobs, shell scripts, or another MCP-capable client. MCP execution requires interactive form elicitation and fails closed in non-interactive clients; plan, draft, dry-run, and direct CLI workflows remain available to automation.

external agent -> hermes-action -> Hermes Agent -> skills/tools/integrations

Why this exists

Coding agents are good at understanding a repository. They should not duplicate your automation stack.

If an agent needs to do something outside its local coding session — research, schedule a job, open a browser workflow, send a message, prepare a social post, use a Hermes skill, or coordinate with a messaging gateway — it can delegate that request to Hermes through this bridge.

Features

• Generic hermes-action run command for one-shot delegation.
• Configurable presets for skills, toolsets, provider/model, source, and max turns.
• Deterministic, language-agnostic safety policy: downgrades execute to request-approval by default, with per-preset trust overrides.
• Explicit --yolo mode for users who intentionally want to bypass bridge-level policy.
• Context file injection with a per-file cap and a configurable aggregate budget (clear error instead of a cryptic E2BIG).
• Automatic large-context delivery: an oversized prompt is handed to Hermes through a secure temp file instead of overflowing the command line.
• Per-run timeouts (configurable, with per-mode defaults) that always reap the Hermes child process.
• Dry-run mode for debugging the exact Hermes command, prompt, and computed prompt size.
• Complete Codex and Claude Code installer: managed skill plus verified global MCP registration.
• MCP capability discovery, cancellable asynchronous jobs, and one-shot two-phase approvals gated on interactive form elicitation, with a privacy-preserving local audit log.
• Local stdio transport plus an opt-in, authenticated Streamable HTTP transport for Tailscale-restricted hosts.
• No project-specific assumptions. All behavior is configured through YAML and CLI flags.

Requirements

• Node.js 20 or newer.
• Hermes Agent installed and available as hermes, or configured with a custom command path.

Check Hermes:

hermes --version

Installation

Install from npm:

npm install -g hermes-action-bridge

Install the skill and MCP registration for Codex and Claude Code, then verify the complete setup:

hermes-action --version

hermes-action install all hermes-action doctor

Global installation is all-or-nothing for the requested agents. If Codex or Claude Code is unavailable, conflicting, or cannot verify its user-scoped stdio MCP entry, the command exits non-zero without installing the requested skills. Use --project for an explicit skill-only setup.

Other distribution channels

Choose the channel that matches the agent surface. Hermes Agent must always be installed and configured on the same computer as the local bridge.

The standalone Agent Skill only teaches an agent when and how to delegate; it does not register an MCP server. Use the npm installer, Claude marketplace, or MCPB channel when MCP tools are required.

Maintainers releasing a new version should follow the distribution runbook, which covers the automated release workflow and the manual directory publication checks.

Codex CLI and the Codex surface in the ChatGPT desktop app share the local Codex MCP configuration on the same host. A general ChatGPT connector is a different distribution target: it requires a remotely reachable MCP server and is not installed by this local stdio bundle.

From source (for development)

git clone https://github.com/TheBlueHouse75/hermes-action-bridge.git

cd hermes-action-bridge npm install npm run build npm link

Quick start

Create a config file in your project:

hermes-action init

Ask Hermes for a safe plan:

hermes-action run --mode plan "Find the best next action from this repository context."

Delegate with a context file:

hermes-action run \

--preset research \ --context ./notes.md \ "Analyze this and return the next concrete action."

Run a dry-run to inspect what will be sent to Hermes:

hermes-action run --dry-run --json "Summarize this project."

Execution modes

• plan: Hermes is instructed to return a plan only. Treat it as open-world because configured tools remain available.
• draft: Hermes produces an artifact only. No external side effects.
• execute: Hermes may execute allowed actions, while still following Hermes' own safety rules.
• request-approval: Hermes prepares an action without executing it. MCP callers should use hermes_prepare and then hermes_approve; the latter opens an interactive client form showing the exact action and executes only after the human confirms it.

The policy guard is deterministic and language-agnostic: in execute mode the bridge switches to request-approval by default — regardless of the prompt's wording or language — unless the preset is explicitly trusted (empty require_approval_for) or --yolo is set. Separately, every effective MCP execute request requires interactive client confirmation, including trusted presets and YOLO. The bridge does not try to infer risk from keywords.

Risk categories below are informational only: the bridge surfaces the ones it recognizes in the prompt envelope to help you and Hermes decide, but they never drive the mode. (Keyword matching is English-biased and trivially evaded, so it must not be a security control.)

• publish_external
• send_message
• send_email
• delete
• payment
• git_push
• credential_change

YOLO mode

YOLO mode is off by default.

hermes-action run --yolo --mode execute "Do the task now."

YOLO bypasses the mode-downgrade policy. It does not bypass the MCP interactive confirmation barrier, Hermes Agent's own safety rules, provider/tool approval prompts, or platform constraints. Direct CLI use remains an explicit trusted-local escape hatch.

Use it only when the caller and environment are trusted.

Trusted presets (direct execution)

By default every execute is downgraded to request-approval — the right default when an agent might act unattended. When *you* are the one asking for an action and want it to run without that extra gate, define a trusted preset: one whose require_approval_for is an empty list. Put it in your user config so it is yours alone and available in every project:

# ~/.config/hermes-action/config.yaml   (user scope, merged before any project config)

presets: act: description: Direct execution of an action I explicitly asked for (trusted). require_approval_for: []

Run actions through it — execute stays execute, with no downgrade and no --yolo needed:

hermes-action run --preset act --mode execute "post the release note to #general"

Keep it safe:

• A trusted preset relaxes the mode-downgrade policy. Effective MCP execution still requires interactive confirmation; direct CLI execution relies on the operator and environment being trusted.
• Keep it in your user config; it never ships in the public package, and the distributed SKILL.md stays cautious for everyone else.
• Reserve trusted presets for actions a human explicitly asks for. Leave the conservative default for anything an agent could trigger on its own.

Configuration

hermes-action loads config in this order:

CLI flags > project .hermes-action.yaml > user config > built-in defaults

User config path:

~/.config/hermes-action/config.yaml

Project config path:

.hermes-action.yaml

Example:

runtime:

adapter: hermes-cli command: hermes

Aggregate budget across all --context files, in bytes (default ~768 KiB).

Raising this above ~896 KiB switches large prompts to temp-file delivery.

max_context_bytes: 786432

Optional overall per-run timeout in seconds. When unset, per-mode defaults

apply: 180s for plan/draft, 600s for execute/request-approval.

timeout_seconds: 600

defaults: mode: plan source: external-agent max_turns: 30 preset: default

presets: default: description: No extra skills or toolsets. Uses the active Hermes profile. skills: [] toolsets: []

research: description: General research and synthesis. skills: [] toolsets: [web, terminal, file]

Per-preset override: relax approvals for a trusted preset.

An empty list never downgrades execute to request-approval.

require_approval_for: []

coding: description: Repository inspection and runtime validation. skills: [developer-assurance-and-validation, runtime-debugging] toolsets: [terminal, file]

policy: yolo: false

Global default; a preset's own require_approval_for takes precedence when set.

require_approval_for:

• publish_external
• send_message
• send_email
• delete
• payment
• git_push
• credential_change

Limits, timeouts, and large context

The bridge builds one prompt envelope (policy header + your request + every block) and hands it to Hermes. Two guardrails keep that reliable:

• Aggregate context budget (runtime.max_context_bytes, default ~768 KiB, plus a fixed 250 KiB per-file cap). Exceeding it fails fast with an actionable message — Context total N bytes exceeds the limit ... Split your handoff or raise runtime.max_context_bytes. — instead of a cryptic E2BIG spawn crash. The default leaves headroom under the operating system's argument-size limit.
• Large-context delivery. If you raise the budget and the envelope grows past ~896 KiB, the adapter writes it to a 0600 temp file and tells Hermes to read that file (injecti