Skills

Skills are pre-built integrations that connect CodeBuddy to external services like GitHub, Jira, databases, cloud providers, and monitoring tools. The SkillManager discovers skills from three sources, handles dependency installation (OS-aware), and injects skill prompts into the agent’s context.

Available skills

Skill	Category	Description
AWS	Cloud	EC2, S3, Lambda, CloudWatch, IAM management
Datadog	Monitoring	Query metrics, logs, traces, and monitors
GitHub	Version Control	Issues, PRs, Actions, branches, releases
Jira	Project Management	Create/update issues, manage sprints
Kubernetes	Infrastructure	Pods, deployments, services, logs
PostgreSQL	Database	SQL queries, schema management, backups
Sentry	Monitoring	Error tracking, issue triage, release health

Additional skills available: Elasticsearch, Email, GitLab, Gmail, Linear, MongoDB, MySQL, Redis, Telegram.

Discovery hierarchy

Skills are loaded from three sources. Higher priority sources override lower ones:

Priority	Source	Path	Scope
1 (highest)	Workspace	`.codebuddy/skills/`	Project-specific skills
2	Global	`~/.codebuddy/skills/`	User-wide skills (all workspaces)
3 (lowest)	Built-in	Bundled with extension	Default skill catalog

Installing a skill

Activate skills through the command palette:

CodeBuddy: Install Skill

Or ask CodeBuddy directly:

Install the GitHub skill so I can manage issues

Dependency management

Skills can declare CLI dependencies (e.g., aws CLI, kubectl, psql). The SkillService handles installation across platforms:

Platform	Package managers
macOS	brew, npm, pip
Linux	apt, snap, npm, pip
Windows	choco, winget, npm, pip

Installation results are cached for 30 seconds to avoid repeated subprocess calls. Architecture-specific binaries (x64, arm64) are selected automatically.

Environment isolation

Skills run with a restricted environment. These sensitive environment variables are blocked:

LD_PRELOAD, DYLD_INSERT_LIBRARIES, PATH, HOME, SHELL, SSH_AUTH_SOCK, GPG_AGENT_INFO, HISTFILE, and others (14 total).

Using skills

Once installed, skills are available automatically. Just ask CodeBuddy to perform actions related to the skill:

Create a Jira ticket for the login bug we just fixed

Show me the Redis cache hit rate from Datadog for the last hour

Deploy the latest image to the staging Kubernetes cluster

Creating custom skills

Create a directory under .codebuddy/skills/ with a SKILL.md file:

.codebuddy/skills/
  my-skill/
    SKILL.md          # Required: skill definition
    scripts/          # Optional: bundled scripts
    install.sh        # Optional: installation script

The SKILL.md uses YAML frontmatter for metadata and Markdown for instructions:

---
name: api-auditor
displayName: API Auditor
description: Audit and test API endpoints for reliability
category: testing
version: 1.0.0
dependencies:
  - name: curl
    check: curl --version
---

# API Auditor

When this skill is active, you audit API endpoints by...

Frontmatter fields

Field	Required	Description
`name`	Yes	Unique identifier
`displayName`	No	Human-readable name for UI
`description`	Yes	When to activate — matched against user requests
`icon`	No	Emoji or icon identifier
`category`	No	Skill category (cloud, database, monitoring, etc.)
`version`	No	Semantic version
`dependencies`	No	CLI tools needed (with check commands)
`config`	No	Configuration fields the skill needs
`auth`	No	Authentication type (api-key, oauth, cli)

See the Skills API reference for a full tutorial on creating custom skills.