PersonaKit makes repeated agent setup inspectable and repeatable.

Turn recurring coding work into a deterministic session contract for role, rules, context, stop points, allowed capabilities, and forbidden actions before handoff.

Start Here

The Practical Problem

The repeated pain is not that agents lack instructions. It is that capable agents, memories, skills, and compressed chat history still need an operator-visible boundary for the work they are about to do.

Setup Lives In Chat

Repeated work modes keep getting rebuilt as long prompts, remembered caveats, and "please be careful, but do not do that other thing" reminders.

Capabilities Need Authority

Host skills and tools can teach useful procedures, but availability is not the same as authorization for the active work mode.

Review Can Drift

A review request can quietly become implementation unless the role, allowed capabilities, forbidden actions, and stop points are inspectable before work starts.

How The Pieces Fit

PersonaKit is the session-start resolver. Prompts describe the task, reusable PersonaKit content defines the pieces, and PersonaKit composes them into a resolved boundary for the handoff.

1. Start The Session

Prompt + Session

The operator asks for work and chooses the PersonaKit session that should govern it.

prompt session id

2. Resolve The Contract

PersonaKit

Deterministically composes persona, directive, kits, intents, essentials, references, and skill authorization into one inspectable contract.

role rules allowed skills stop points

3. Work Under It

Agent + Skills

The person at the controls or the harness passes the exported contract into the coding agent or host surface. Skills and tools can help, but only when the contract authorizes them.

harness skills tools

PersonaKit Studio sits on the same contract model as a visual inspection and pack-maintenance surface. The CLI and static example roots remain the primary path for validating, inspecting, and exporting contracts. Learn introduces the guided ReleaseDesk lab once the basic contract model is in place.

Run One Contract

Try The Demo

cd Site/public/examples/swift-cli-maintenance
personakit validate --root personakit-root
personakit contract --root personakit-root --session cli-maintenance
personakit export --root personakit-root --session cli-maintenance

What It Proves

  • The authored root is valid.
  • The session resolves deterministically.
  • Code editing is authorized for this work mode.
  • Deployment and autonomous loops are forbidden.
  • The handoff context is visible before work starts.
Run the first contract

Think In Contracts

Session, persona, directive, kits, intents, essentials, references, and skill authorization resolving into an operating contract for an agent.

Example Roots

After the Start Here guide, use these roots as practical patterns for recurring work modes.

CLI Maintenance

The happy path for bounded implementation with one authorized capability and clear stop points.

Open example

MCP Grounding

How MCP-aware agents should resolve PersonaKit context before choosing host-local tools.

Open example

Review Session

A read-only review contract that produces findings without becoming the implementer.

Open example

Pack Authoring

A compact authoring contract for creating PersonaKit content with dry-runs, reuse, and validation.

Open example

Use It When The Work Mode Repeats

Good Fit

Use it at the start of a repeated session when a prompt plus skills is not enough: role, rules, allowed capabilities, stop points, and proof need to be inspectable before work begins.

Bad Fit

Skip it when you only need a one-off prompt, a skill or tool inside an already-running conversation, a place to track tasks, or exploratory planning before the work mode is clear.

Check fit before authoring