okr-manager

OKR lifecycle management for AI agents.

Repo: fractalmind-ai/okr-manager-skillLanguage: Skill (Markdown-based) Status: Stable Install: npx openskills install fractalmind-ai/okr-manager-skill

What It Does

okr-manager gives AI agents the ability to autonomously manage Objectives and Key Results:

• Create OKRs: With quality gates ensuring measurability and accountability
• Track progress: Status markers, dependency chains, priority levels
• Heartbeat audit: Automatic OKR review during agent heartbeats
• Report status: Formatted progress reports for humans

Quality Gates

Every OKR must pass quality checks before creation:

• Objective is outcome-oriented (not task-oriented)
• Success criteria are measurable and binary-verifiable
• Each KR has status markers (PENDING, IN PROGRESS, COMPLETE)
• Dependencies between KRs are explicit
• Deliverables are specified
• Priority and owner are assigned

Usage

npx openskills read okr-manager

The skill guides the agent through OKR operations:

## Create a new OKR

### Objective
Build a complete documentation site for the protocol.

### Key Results
KR1: Architecture docs covering all 9 modules — PENDING
KR2: SDK API reference with examples — PENDING
KR3: Deploy to GitHub Pages — PENDING

Configuration

Supports multiple languages and custom markers:

language: en  # or zh, ja
status_markers:
  pending: "PENDING"
  in_progress: "IN PROGRESS"
  complete: "COMPLETE"
priority_markers:
  p0: "P0"
  p1: "P1"

Where It Fits

okr-manager is independent — it doesn't depend on other skills but is commonly used with:

• agent-manager: OKR review during heartbeat cycles
• team-manager: Team-level OKR tracking
• team-chat: OKR progress shared via team messages

Related Components

• agent-manager — heartbeat integration
• team-manager — team-level goals