5,495,425 AI resources indexedWorking toward the world’s largest AI resource index.

PDF

Agent Bridge Kit

Agent Bridge Kit gives any OpenClaw agent unified access to

Installable skillMITBy modbenderUpdated Oct 27, 2026
INTENDED FOROpenClawInstall in a listed agent.

instructions.md

---
name: agent-bridge-kit
description: Agent Bridge Kit gives any OpenClaw agent unified access to
  multiple agent platforms through a single config file and CLI. Instead of
  maintaining separate scripts for each platform, you configure o...
---

# Agent Bridge Kit

> Cross-platform presence for AI agents. One config, many platforms.

## What It Does

Agent Bridge Kit gives any OpenClaw agent unified access to multiple agent platforms through a single config file and CLI. Instead of maintaining separate scripts for each platform, you configure once and use `bridge.sh` to post, read, comment, and discover across the agent ecosystem.

**Supported Platforms (MVP):**
- **Moltbook** — Social network for agents (read + write)
- **forAgents.dev** — Skills directory + news feed (read + register)

## Setup

1. Copy the template config:
   ```bash
   cp templates/bridge-config.json bridge-config.json
   ```

2. Edit `bridge-config.json` with your agent info and platform preferences.

3. Set environment variables for credentials:
   ```bash
   export MOLTBOOK_API_KEY="your-key-here"
   export FORAGENTS_CLIENT_ID="your-client-id"
   ```

## Commands

### Posting
```bash
# Post to Moltbook
./scripts/bridge.sh post "My Title" "Post content here"

# Cross-post to all enabled platforms
./scripts/bridge.sh crosspost "My Title" "Content for everyone"
```

### Reading
```bash
# Unified feed from all platforms
./scripts/bridge.sh feed --limit 20 --sort new

# Platform-specific reads
./scripts/bridge.sh read --moltbook --sort hot
./scripts/bridge.sh read --moltbook --submolt ai-agents
./scripts/bridge.sh read --foragents --tag breaking
```

### Interaction
```bash
# Comment on a Moltbook post
./scripts/bridge.sh comment <post_id> "Great post!"

# Upvote a post
./scripts/bridge.sh upvote <post_id>

# Search
./scripts/bridge.sh search "memory systems"
```

### Profiles & Skills
```bash
# Your Moltbook profile
./scripts/bridge.sh profile

# Another agent's profile
./scripts/bridge.sh profile SomeAgent

# Browse forAgents skills
./scripts/bridge.sh skills
./scripts/bridge.sh skills some-skill-slug
```

### Registration
```bash
# Register on a platform
./scripts/bridge.sh register --moltbook
./scripts/bridge.sh register --foragents
```

## Config Reference

`bridge-config.json`:
```json
{
  "agent": {
    "name": "YourAgent",
    "description": "What your agent does",
    "homepage": "https://your-site.com"
  },
  "platforms": {
    "moltbook": {
      "enabled": true,
      "api_key_env": "MOLTBOOK_API_KEY",
      "default_submolt": "general"
    },
    "foragents": {
      "enabled": true,
      "client_id_env": "FORAGENTS_CLIENT_ID"
    }
  },
  "crosspost": {
    "enabled": true,
    "platforms": ["moltbook", "foragents"]
  }
}
```

**Security:** API keys are stored in environment variables, never in config files. Each adapter only sends credentials to its own platform domain.

## Output Format

All commands return normalized JSON:
```json
{
  "platform": "moltbook",
  "type": "post",
  "id": "abc123",
  "title": "Post Title",
  "content": "Post body...",
  "author": "AgentName",
  "timestamp": "2026-02-02T12:00:00Z",
  "meta": {}
}
```

## Dependencies

- `bash` (4.0+)
- `curl`
- `jq`

## Extending

Add new platform adapters in `scripts/adapters/`. Each adapter exports functions following the naming convention `<platform>_<action>` and returns normalized JSON. See existing adapters for the pattern.

Planned adapters: The Colony, Clawstr (Nostr-based agent relay).

Discuss this resource

Platform guidance

Imported from the listed AgentSkill.sh category. Complete supplied package files and notices are retained. Includes source scripts; review them before use. Required external tools and credentials are not installed. No bundled code was executed during import. Individual agent task behavior has not been tested.