Delegate independent tasks to specialized agents working concurrently with isolated context.
Hive Project Management
hive-project-management
Manage Agent Hive projects using AGENCY.md files. Use this skill when creating, updating, or working with AGENCY.md project files, managing project metadata (status, owner, blocked, priority), marking tasks complete, adding agent notes, or understanding the Agent Hive shared memory system.
SKILL.md
Full skill instructions
Hive Project Management
This skill guides you in managing Agent Hive projects through AGENCY.md files - the shared memory primitive that enables coordination between humans and AI agents.
AGENCY.md Structure
Every project has an AGENCY.md file with YAML frontmatter:
---
project_id: unique-identifier
status: active # active, pending, blocked, completed
owner: null # null or agent name (e.g., "claude-3.5-sonnet")
last_updated: 2025-01-15T10:30:00Z
blocked: false
blocking_reason: null
priority: high # low, medium, high, critical
tags: [feature, backend]
dependencies: # Optional dependency tracking
blocked_by: [] # Projects that must complete first
blocks: [] # Projects this blocks
parent: null # Parent project ID
related: [] # Related projects
---
# Project Title
## Objective
What this project aims to achieve.
## Tasks
- [ ] Task 1
- [ ] Task 2
- [x] Completed task
## Agent Notes
- **2025-01-15 10:30 - Claude**: Started work on Task 1
Frontmatter Fields Reference
| Field | Type | Description |
|---|---|---|
project_id | string | Unique identifier for the project |
status | enum | active, pending, blocked, completed |
owner | string/null | Agent name currently working, or null if unclaimed |
last_updated | ISO timestamp | When the file was last modified |
blocked | boolean | Set to true if human intervention needed |
blocking_reason | string/null | Explanation of what's blocking |
priority | enum | low, medium, high, critical |
tags | array | Organizational labels |
dependencies | object | Dependency relationships (see below) |
Dependencies Object
dependencies:
blocked_by: [project-a, project-b] # Must wait for these to complete
blocks: [project-c] # These projects wait for this one
parent: parent-project-id # Hierarchical relationship
related: [sibling-project] # Informational links
Working with Projects
Claiming a Project
Before starting work on a project:
- Check that
ownerisnull(unclaimed) - Check that
statusisactive - Check that
blockedisfalse - Verify no unresolved
dependencies.blocked_by - Set
ownerto your agent name
owner: "claude-sonnet-4"
Updating Progress
When completing tasks:
- Mark tasks with
[x]in the markdown content - Update
last_updatedtimestamp - Add a note to the "Agent Notes" section
Adding Agent Notes
Format notes with timestamp and agent name:
## Agent Notes
- **2025-01-15 14:30 - claude-sonnet-4**: Completed research phase. Found 5 sources.
- **2025-01-15 10:00 - claude-sonnet-4**: Starting work on research tasks.
Setting Blocked Status
If you need human intervention:
blocked: true
blocking_reason: "Need API credentials for external service"
Completing a Project
When finished:
- Mark all tasks complete
- Set
status: completed - Set
owner: null - Add final notes
Releasing Without Completing
If handing off to another agent:
- Set
owner: null - Add notes about current state
- Leave
statusunchanged
Creating New Projects
To create a new project:
- Create directory:
projects/your-project-name/ - Create
AGENCY.mdwith proper frontmatter - Define clear objectives and tasks
- Set initial
status: pendingorstatus: active
Best Practices
- Always claim before working - Set
ownerto prevent conflicts - Update frequently - Keep
last_updatedcurrent - Be descriptive in notes - Future agents need context
- Use blocking correctly - Only set
blocked: truefor external dependencies - Manage dependencies - Keep
blocked_by/blocksaccurate - Release when done - Set
owner: nullwhen finished or handing off - Follow priority - Work on
criticalandhighpriority first
Common Patterns
Research then Implementation
## Tasks
### Research Phase
- [x] Find sources
- [x] Summarize findings
### Implementation Phase
- [ ] Design solution
- [ ] Implement feature
- [ ] Test changes
Phased Rollout
Use dependencies.blocked_by to sequence phases:
# Phase 2 project
dependencies:
blocked_by: [phase-1-project]
Parent-Child Projects
# Child project
dependencies:
parent: main-feature-project
More Other skills
brand-guidelines
Applies Anthropic's official brand colors and typography to any sort of artifact that may benefit from having Anthropic's look-and-feel. Use it when brand colors or style guidelines, visual formatting, or company design standards apply.
theme-factory
Toolkit for styling artifacts with a theme. These artifacts can be slides, docs, reportings, HTML landing pages, etc. There are 10 pre-set themes with colors/fonts that you can apply to any artifact that has been creating, or can generate a new theme on-the-fly.
internal-comms
A set of resources to help me write all kinds of internal communications, using the formats that my company likes to use. Claude should use this skill whenever asked to write some sort of internal communications (status reports, leadership updates, 3P updates, company newsletters, FAQs, incident reports, project updates, etc.).
Fundamental design principles for writing maintainable, testable Python code.
responsive-design
Modern responsive layouts using container queries, fluid typography, CSS Grid, and mobile-first strategies.
Generate Playwright tests from scenarios using interactive browser exploration and validation.
Automated website exploration and test case generation using Playwright.
Modern React state management with Redux Toolkit, Zustand, Jotai, and React Query for every state category.
gdpr-data-handling
GDPR-compliant data handling with consent management, data subject rights, and privacy controls.
Python error handling patterns including input validation, exception hierarchies, and partial failure handling. Use when implementing validation logic, designing exception strategies, handling batch processing failures, or building robust APIs.
refactor-plan
Detailed multi-file refactor planning with safety checks and rollback guidance.
