> ## Documentation Index
> Fetch the complete documentation index at: https://docs.sureops.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Skills

> Reusable tool bundles that extend what agents can do — and how to configure them.

Skills are reusable bundles of tools and instructions that extend what an agent can do during an incident. Rather than giving every agent access to every possible tool, skills let you attach specific capabilities to specific agents, keeping each agent focused and auditable.

## What a skill contains

A skill has two parts:

* **Tool hints**: references to specific tools (MCP endpoints, API wrappers) the agent can call when using this skill
* **Instructions**: guidance for the agent on when and how to use the tools in this skill

When an agent runs, it receives the combined set of tools and instructions from all skills attached to it.

***

## Built-in skills

### landscape-context

The `landscape-context` skill gives agents access to your service topology — the graph of services, dependencies, ownership, and deployment information built from your `.sureops/` contract and telemetry discovery.

**Available to**: Diagnosis Specialist, Incident Commander, Problem Investigator, Fix-PR Agent

**Tools included:**

| Tool                            | What it does                                                                            |
| ------------------------------- | --------------------------------------------------------------------------------------- |
| `landscape_get_service_context` | Returns ownership, runbook, tier, dependencies, and deployment info for a named service |
| `find_blast_radius`             | Given a service, returns all downstream services that could be affected                 |
| `find_upstream_dependencies`    | Returns all upstream services a given service depends on                                |
| `find_aws_dependencies`         | Returns AWS resources associated with a service (EC2, RDS, EKS, etc.)                   |
| `lookup_alert`                  | Maps an alert's labels (service name, namespace, pod) to the canonical service entry    |

The landscape-context skill is automatically attached to the agents above and does not require manual configuration. You can disable it per-agent if needed from **Settings → AI Agents → \[agent] → Skills**.

### GitHub agent skills

GitHub skills give agents read access to your code repositories for diagnosis and the ability to open pull requests for fix-PR and postmortem workflows.

**Available to**: Fix-PR Agent, Problem Investigator

**Tools included:**

* `github_read_file` — Read a specific file from a repository
* `github_search_code` — Search for patterns in repository code
* `github_create_pr` — Open a pull request (used by Fix-PR Agent)
* `github_get_pr_diff` — Read a PR diff (used by Pre-mortem Agent)

GitHub skills require a connected GitHub integration. See the Integrations section for setup.

***

## Attaching skills to agents

<Steps>
  <Step title="Open agent settings">
    Go to **Settings → AI Agents** and select the agent.
  </Step>

  <Step title="Open the Skills tab">
    On the agent configuration card, select the **Skills** tab.
  </Step>

  <Step title="Toggle skills">
    Enable or disable skills for this agent. Each skill shows a description and the list of tools it provides.
  </Step>

  <Step title="Save">
    Changes take effect on the next incident.
  </Step>
</Steps>

***

## Custom skills

You can define custom skills that reference your own MCP endpoints. This is useful for integrating proprietary tools — internal runbook APIs, custom dashboards, or in-house automation platforms — without building a new integration.

A custom skill requires:

* A name and description
* A list of tool hints that reference MCP endpoint names
* Instructions for the agent on when to call the tools

Custom skills are configured via **Settings → AI Agents → Skills → New skill**. The tool hints must match the names of tools exposed by a connected MCP integration.

***

## How skills interact with agent instructions

When an agent runs, it receives:

1. The base agent instructions (sureops-defined; not customer-configurable)
2. Tool hints from all attached skills
3. Per-skill instructions
4. Custom agent instructions you've added via **Settings → AI Agents → \[agent] → Instructions** (appended, not replacing the base)

The agent uses these to decide which tools to call and how to reason about their results. Skills are additive — they extend, not replace, the base agent behavior.

***

## Auditing skill tool calls

Every tool call made by an agent via a skill is recorded in the incident timeline. The entry shows:

* The tool name (from the skill)
* The input parameters
* The output (or error)
* The timestamp

This gives you a complete record of what each agent queried during the incident, regardless of which skill provided the tool. See [Confidence & Audit Trace](/next/operating-modes/confidence-and-audit-trace) for more on reading the incident timeline.