prime

Load project context and display active status from workspace metadata.

2|1|Updated Jan 27, 2026
One-click install
npx skills add https://github.com/lucidlabs-hq/agent-kit --skill prime-lucidlabs-hq
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: prime
Source: https://github.com/lucidlabs-hq/agent-kit/tree/main/.claude/skills/prime
Command: npx skills add https://github.com/lucidlabs-hq/agent-kit --skill prime-lucidlabs-hq

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automatically load the current project context and surface the active status so teams can resume work without manual setup.

Core Features & Use Cases

  • Context detection and header rendering: Detect whether the workspace is upstream or downstream and render a standardized CONTEXT header.
  • Active project discovery: Read active_project.name and active_project.prd from the context metadata; support .claude/PROJECT-CONTEXT.md if available.
  • Boot/Handoff support: Prepare the session boot sequence and hand off to downstream skills without requiring a manual session reset.
  • Safety & scope guardrails: Enforce Agent Kit command boundaries to prevent unintended changes to upstream templates.

Quick Start

At session start, load the current project context and display the active status to orient your work.

Frequently Asked Questions about prime

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I load project context and show active status at session startup?

Use a session-start context detection tool to read workspace state, locate .claude/PROJECT-CONTEXT.md, parse active_project metadata, and render a standardized status header. This automatically loads project context to orient work without manual setup.

What is project context detection and how does it work for upstream and downstream workspaces?

Project context detection reads the repository type to determine whether a workspace is upstream or downstream, then renders a standardized CONTEXT header. It parses active_project.name and active_project.prd metadata to surface the current development state.

How do I prevent unintended changes to upstream templates when loading workspace context?

Use an agent kit with built-in safety scope guardrails that enforce strict command boundaries during context loading. These guardrails prevent unintended modifications to upstream templates while parsing and rendering the active project status.

Can I resume work without a manual session reset after loading project context?

Yes, context detection tools support boot and handoff sequences that prepare the session and pass control to downstream skills automatically. This eliminates manual session resets by parsing active_project metadata and rendering boot prompts directly.

Does context detection require a .claude/PROJECT-CONTEXT.md file to function?

The tool supports .claude/PROJECT-CONTEXT.md if available but does not strictly require it to function. It reads workspace state, detects repository type, and parses active_project metadata to render the status header regardless of file presence.