Skip to main content
Skills

AI-Powered Automation Skills

Skills are AI-driven automation units that define how Akmatori investigates and responds to incidents. Each skill contains operator-authored instructions plus generated gateway examples for assigned tools.

What is a Skill?

A skill is defined by a SKILL.md file that contains YAML frontmatter with metadata and markdown content with instructions for the AI agent. Akmatori appends an auto-generated Assigned Tools section whenever tool assignments change, so agents get current gateway_call(...) examples without hand-written credentials or import paths.

Skills follow the open Agent Skills specification, making them portable across compatible AI platforms.

SKILL.md Format

Here's an example skill definition:

SKILL.md
---
name: investigate-high-cpu
description: Investigate and diagnose high CPU usage alerts
version: 1.0.0
---

# Investigation Instructions

When receiving a high CPU alert, follow these steps:

1. Confirm the affected host and service from the alert payload
2. Use the assigned SSH and monitoring tools through gateway_call examples
3. Check CPU, memory, recent deploys, and relevant runbooks
4. Summarize findings and recommend next steps

## Expected Output

Provide a structured report with:
- Root cause analysis
- Affected services
- Recommended remediation steps

Skill Properties

nameRequired

Unique identifier for the skill

descriptionRequired

Brief description of what the skill does

versionOptional

Semantic version number (default: 1.0.0)

toolsOptional

Assigned through the UI/API; Akmatori writes the generated Assigned Tools section rather than storing tool IDs in frontmatter

Assigned Tools

Tool access is managed outside the hand-written prompt. Assign tool instances to a skill in the UI or with PUT /api/skills/:name/tools, and Akmatori regenerates SKILL.md with the logical tool name, parameter schema, and executable gateway examples.

generated section
## Assigned Tools

Your assigned tools are listed below with full parameter schemas and gateway_call examples.

### ssh-production (logical_name: "prod-ssh", type: ssh)

Usage (via gateway_call):

```python
gateway_call("ssh.execute_command", {"command": "uptime"}, "prod-ssh")
gateway_call("ssh.test_connectivity", {}, "prod-ssh")
```

Use `execute_script` for multi-step scripts with built-in `gateway_call()`.

Best Practices

Be Specific

Write clear, specific instructions for the AI agent. Avoid ambiguity.

Add Context Files

Upload runbooks and documentation as context files for better analysis.

Test Thoroughly

Run skills manually before enabling automatic triggers.