An AI-driven scouting system for logging field observations, pests, and equipment issues directly into FarmOS.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install farmos-observations
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install farmos-observations using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is FarmOS Observations?
This skill provides an AI-powered quick capture system designed specifically for field-based reporting and agricultural management. It enables users to document pests, diseases, weeds, and crop conditions through natural language inputs, photos, or voice notes. By leveraging Openclaw Skills, the agent intelligently categorizes reports, assesses severity based on linguistic cues, and integrates these findings directly with existing FarmOS workflows.
Beyond simple data entry, it acts as an intelligent field assistant that recognizes common agricultural issues, specifically tailored for regional challenges like Indiana-specific pests and diseases. The skill ensures data completeness by tracking observation counts and providing summary statistics through dedicated dashboard endpoints, making it an essential tool for modern precision agriculture and team-wide field coordination.
FarmOS Observations Use Cases
- Identifying and reporting pest or disease outbreaks in specific field locations using natural language.
- Logging equipment failures or maintenance needs discovered during scouting for immediate escalation.
- Documenting weather damage or soil erosion after major storm events to track long-term impact.
- Tracking crop growth stages and general field health over time to optimize treatment schedules.
- Creating field-based scouting records directly from photos taken on a mobile device.
How FarmOS Observations Works
- The user provides a natural language trigger, such as "beans look rough in field 12," or uploads a photo representing an observation.
- Openclaw Skills analyze the input to automatically detect the field ID, observation type (e.g., weed, pest, equipment), and urgency level.
- If critical information is missing, the agent utilizes a smart follow-up logic to ask targeted questions about scope, severity, or location without over-interrogating the user.
- Validated data is sent to the FarmOS API via a multipart/form-data POST request, including any media attachments and AI-generated descriptions.
- The skill offers proactive post-creation actions, such as generating work orders in other modules or suggesting follow-up scouting trips.
FarmOS Observations Setup
To get started, ensure you have the necessary authentication script to generate a JWT for the FarmOS backend. Run the following command to retrieve your token:
TOKEN=$(~/clawd/scripts/farmos-auth.sh manager)
Configure your environment to point to the FarmOS API base. When interacting with the observations endpoints, include the token in your authorization headers as shown below:
curl -H "Authorization: Bearer $TOKEN" http://100.102.77.110:8008/api/observations
Note: Tokens last for 15 minutes. If you receive a 401 response, simply request a new token using the auth script.
FarmOS Observations Data Schema & Taxonomy
Observations are organized by type and urgency to facilitate rapid decision-making. The system utilizes the following schema for data organization:
| Attribute | Description |
|---|---|
| observation_type | Categorized as pest, disease, weed, weather_damage, equipment_issue, soil, or crop_condition |
| severity | Qualitative assessment (low, medium, high) mapped from user language |
| urgency_score | A 1-10 scale used for escalation (scores 7+ trigger immediate alerts) |
| field_id | Numeric identifier linking the observation to a specific farm location |
| photo | Image file attachment archived with the observation for visual proof |
| entities | Extracted metadata including specific pest names, equipment IDs, and cardinal directions |
FarmOS Observations Advanced Features
- Smart identification of regional pests, diseases, and weed species from common names and colloquialisms.
- Automated urgency escalation for high-severity issues like chemical exposure, structural failures, or rapid disease spread.
- Cross-module context linking Openclaw Skills to weather patterns, equipment maintenance history, and existing task lists.
- Enhanced image understanding that extracts detailed diagnostic descriptions from photos to populate observation records.
- Proactive pattern detection that alerts managers when similar observations appear across multiple fields within a short timeframe.
SKILL.md
Loading
Related Openclaw Skills
FarmOS Land Portfolio
A specialized management tool for tracking agricultural land ownership, lease agreements, landlord contacts, and financial payment schedules.
FarmOS Finance
A professional financial management skill for agricultural operations to track costs, project cash flows, and analyze breakeven points with admin-level security.
FarmOS Equipment Management Skill
A comprehensive tool for monitoring farm equipment status, tracking maintenance schedules, and accessing service manuals using AI-driven insights.
AgResource Newsletter Scraper
A sophisticated automation tool that scrapes, analyzes, and summarizes AgResource grain marketing newsletters for actionable agricultural insights.
FarmOS Task Manager
A powerful tool for Openclaw Skills to manage agricultural work orders, track task assignments, and monitor farm productivity in real-time.
FarmOS Weather
A specialized skill for querying real-time weather, forecasts, and agronomic metrics like GDD for specific farm fields.
Featured*
