By olakai-ai
Official skills for Olakai AI observability platform, providing monitoring, analysis, and governance capabilities for AI agents.
Install the CLI, authenticate, and send your first monitored event. AUTO-INVOKE when: CLI is not installed, user is not authenticated, user asks about getting started with Olakai, user wants to try Olakai, user mentions they don't have an account yet, or when other skills fail due to missing prerequisites. TRIGGER KEYWORDS: get started, setup, install olakai, olakai account, sign up, signup, new to olakai, try olakai, olakai onboarding, first agent, getting started, no account, create account, register. DO NOT load for: users who are already authenticated and have agents set up (use olakai-new-project, olakai-integrate, or olakai-troubleshoot instead).
Add Olakai to existing AI code with minimal changes — wrap your LLM client, add custom metrics, and validate the integration. AUTO-INVOKE when user wants to: add monitoring to existing code, integrate observability into a working agent, add tracking to LLM calls, wrap OpenAI/Anthropic clients, add analytics to working AI code, or retrofit monitoring to production AI. TRIGGER KEYWORDS: olakai, add monitoring, integrate monitoring, wrap client, observability, SDK integration, @olakai/sdk, olakai-sdk, existing agent, existing code, add tracking, analytics, telemetry, retrofit monitoring, existing LLM, existing OpenAI, existing Anthropic. DO NOT load for: creating new agents from scratch (use olakai-new-project), troubleshooting issues (use olakai-troubleshoot), or generating reports (use olakai-reports).
Renamed to olakai-monitor-local-coding-agent. This skill now covers all three local coding agents (Claude Code, Codex CLI, Cursor) under one entry point. Existing plugin installs that reference olakai-monitor-claude-code continue to resolve here so they don't break — but the canonical, up-to-date guidance lives in olakai-monitor-local-coding-agent. Load that skill instead.
Diagnose and repair Olakai monitoring for a local coding tool you already set up — Claude Code, OpenAI Codex CLI, or Cursor. Drives `olakai monitor list`, `olakai monitor doctor [--fix]`, and `olakai monitor repair` to self-heal hooks-based monitoring (no events, missing KPIs, broken/deleted agent, drifted config). For first-time setup use olakai-monitor-local-coding-agent instead. AUTO-INVOKE when user says: my coding-tool monitoring isn't working, no events from Claude Code / Codex / Cursor, is monitoring on / working, check my olakai monitoring, what's monitored on this machine, monitor doctor, monitor repair, monitor list, fix my monitoring, my monitored agent disappeared / 404, hooks stopped firing, re-link my monitoring key. TRIGGER KEYWORDS: olakai monitor doctor, olakai monitor repair, olakai monitor list, monitor not working, no events claude code, no events codex, no events cursor, fix monitoring, repair monitoring, agent 404, agent missing, hooks stopped firing, drift, registry, agents mine, where am i monitoring, is my monitoring working, self-heal monitoring. DO NOT load for: first-time setup of a coding tool (use olakai-monitor-local-coding-agent), instrumenting your own agent's SDK code (use olakai-integrate), generic SDK / KPI / event troubleshooting unrelated to a coding tool (use olakai-troubleshoot).
Set up and self-heal Olakai monitoring for the coding tool you are using — Claude Code, OpenAI Codex CLI, Cursor, Google Gemini CLI, or Antigravity CLI. Installs hooks, creates the agent record, and explains how to enrich events with KPIs. This is the skill for "monitor my coding tool itself" (not for instrumenting your own agent's source code with the SDK — that is olakai-integrate). AUTO-INVOKE when user wants to: monitor Claude Code / Codex / Cursor / Gemini CLI / Antigravity CLI sessions, monitor THIS coding tool, add observability to a local coding agent, track my own coding-assistant usage, set up olakai monitoring in this workspace, see what is being monitored on this machine, check if monitoring is working, or enable / repair hooks-based monitoring for any local coding agent. TRIGGER KEYWORDS: olakai monitor, monitor my coding tool, monitor this tool, monitor claude code, monitor codex, monitor cursor, monitor gemini cli, monitor antigravity, codex cli, cursor hooks, gemini cli, gemini-cli hooks, antigravity cli, antigravity, agy, local coding agent, local agent monitoring, olakai hooks, olakai monitor init, olakai monitor list, olakai monitor doctor, olakai monitor repair, monitor workspace, track sessions, is my monitoring working, monitoring not working, no events from claude code, claude code monitoring, codex monitoring, cursor monitoring, agents mine, where am i monitoring. DO NOT load for: instrumenting your own agent's SDK code (use olakai-integrate), creating agents from scratch with custom code (use olakai-new-project), generic SDK / KPI / event troubleshooting unrelated to a coding tool (use olakai-troubleshoot).
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Official skills for integrating AI agents with Olakai - the enterprise AI observability and governance platform.
These skills follow the Agent Skills Standard and work with Claude Code, Cursor, VS Code Copilot, and other compatible AI coding assistants.
# List available skills
npx add-skill olakai-ai/olakai-skills --list
# Install a specific skill
npx add-skill olakai-ai/olakai-skills/olakai-get-started
npx add-skill olakai-ai/olakai-skills/olakai-new-project
npx add-skill olakai-ai/olakai-skills/olakai-integrate
npx add-skill olakai-ai/olakai-skills/olakai-troubleshoot
npx add-skill olakai-ai/olakai-skills/olakai-reports
# Clone to user-level skills directory
git clone https://github.com/olakai-ai/olakai-skills ~/.claude/skills/olakai-skills
# Or clone to project-level
git clone https://github.com/olakai-ai/olakai-skills .claude/skills/olakai-skills
| Skill | Description |
|---|---|
| olakai-get-started | Install the CLI, authenticate, and send your first monitored event |
| olakai-new-project | Build a new AI agent project from scratch with full observability |
| olakai-integrate | Add Olakai to existing AI code with minimal changes |
| olakai-troubleshoot | Diagnose and fix issues with events, KPIs, or SDK integration |
| olakai-reports | Generate usage summaries, KPI trends, ROI reports from the terminal |
| olakai-monitor-local-coding-agent | Set up hooks-based monitoring for local coding agents — Claude Code, Codex CLI, Cursor (olakai monitor init --tool <tool>) |
| olakai-monitor-doctor | Self-heal a coding tool's monitoring — olakai monitor list / doctor [--fix] / repair |
The olakai-expert agent combines all four skills into a single specialist:
| Agent | Description |
|---|---|
| olakai-expert | Full Olakai integration specialist - creates agents, adds monitoring, troubleshoots issues, generates reports |
Before using these skills, ensure you have:
npm install -g olakai-cliolakai loginAlways validate your Olakai integration by generating a test event and inspecting it:
# 1. Run your agent/trigger LLM call
# 2. Fetch the event
olakai activity list --agent-id AGENT_ID --limit 1 --json
olakai activity get EVENT_ID --json | jq '{customData, kpiData}'
# 3. Validate:
# - customData has all expected fields
# - kpiData shows NUMBERS (not strings like "MyVariable")
# - kpiData shows VALUES (not null)
# CLI Authentication
olakai login
olakai whoami
# View Activity
olakai activity list --limit 10
olakai activity get EVENT_ID --json
olakai activity sessions --agent-id ID # Session decoration status
# Manage Agents (includes API key generation)
olakai agents list
olakai agents create --name "Agent Name" --with-api-key --json
olakai agents get AGENT_ID --json | jq '.apiKey'
# Manage KPIs
olakai kpis list --agent-id ID
olakai kpis create --formula "Variable" --agent-id ID
# Manage Custom Data
olakai custom-data list
olakai custom-data create --name "Field" --type NUMBER
TypeScript:
import { OlakaiSDK } from "@olakai/sdk";
const olakai = new OlakaiSDK({ apiKey: process.env.OLAKAI_API_KEY! });
await olakai.init();
const openai = olakai.wrap(new OpenAI({ apiKey }), { provider: "openai" });
Python:
from olakaisdk import olakai_config, instrument_openai
olakai_config(os.getenv("OLAKAI_API_KEY"))
instrument_openai()
For automatic skill activation (recommended for heavy Olakai users):
# Install activation hooks for improved auto-invocation rate
cp hooks/skill-activator.sh ~/.claude/hooks/
chmod +x ~/.claude/hooks/skill-activator.sh
See hooks/README.md for full setup instructions.
See AGENTS.md for full compatibility information with various AI coding agents.
MIT License - see LICENSE for details.
Contributions welcome! Please read our contributing guidelines before submitting PRs.
Built with love by Olakai
npx claudepluginhub olakai-ai/olakai-skillsOlakai - measure AI ROI, govern risk, and control costs. Create agents with analytics, add tracking to existing code, troubleshoot, and generate reports
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
This skill should be used when users need to generate ideas, explore creative solutions, or systematically brainstorm approaches to problems. Use when users request help with ideation, content planning, product features, marketing campaigns, strategic planning, creative writing, or any task requiring structured idea generation. The skill provides 30+ research-validated prompt patterns across 14 categories with exact templates, success metrics, and domain-specific applications.
Develop, test, build, and deploy Godot 4.x games with Claude Code. Includes GdUnit4 testing, web/desktop exports, CI/CD pipelines, and deployment to Vercel/GitHub Pages/itch.io.
A growing collection of Claude-compatible academic workflow bundles. Covers scientific figures, manuscript writing and polishing, reviewer assessment, citation retrieval, data availability, paper reading, literature search, response letters, paper-to-PPTX conversion, and evidence-grounded Chinese invention patent drafting. Rules are organized as reusable skill folders with explicit workflows and quality checks.
Upstash Context7 MCP server for up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your LLM context.
Consult multiple AI coding agents (Gemini, OpenAI, Grok, Perplexity, plus codex, antigravity, and grok CLIs when installed) to get diverse perspectives on coding problems