archon

Orchestrate multi-session campaign execution with git checkpoints and verification steps.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/mferris77/SpringBoard --skill archon-mferris77
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: archon
Source: https://github.com/mferris77/SpringBoard/tree/main/vscode-citadel-harness/skills/archon
Command: npx skills add https://github.com/mferris77/SpringBoard --skill archon-mferris77

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Archon solves the problem of managing large, multi-session engineering work that would otherwise lose context, drift in scope, or suffer from inconsistent quality by providing persistent campaign state, structured decomposition, and enforced verification between sessions.

Core Features & Use Cases

  • Strategic Decomposition: Breaks a direction into 3–8 machine-verifiable phases with explicit end conditions.
  • Delegation & Orchestration: Spawns and coordinates sub-agents with full context injection and logs delegation lifecycle events.
  • Continuation & Checkpointing: Creates git stash checkpoints, writes Continuation State, and can daemonize longer campaigns for automated continuation.
  • Quality & Safety Gates: Runs quality spot-checks, regression guards, anti-pattern scans, and verification steps (typechecks, tests) before advancing.
  • Recovery & Telemetry: Provides rollback procedures, Decision Logs, and telemetry hooks for campaign lifecycle tracking.
  • Use Case: Long-running refactors, cross-system feature builds, or complex migrations that require staged work, review, and persistent state across sessions.

Quick Start

Invoke Archon with a short direction like "Refactor the auth flow into modular services" or run /archon to resume an active campaign.

Frequently Asked Questions about archon

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

FAQPage Schema
How do I manage long-running engineering work across multiple sessions without losing context?

Multi-session engineering work can be orchestrated by decomposing it into 3–8 verifiable phases with explicit end conditions, delegating to sub-agents, and persisting Continuation State with git checkpoints. This prevents context loss and scope drift across sessions.

What is the best way to decompose a complex codebase migration into machine-verifiable phases?

Strategic decomposition breaks a direction into 3–8 machine-verifiable phases with explicit end conditions. Each phase must pass quality spot-checks, regression guards, typechecks, and tests before advancing, ensuring complex migrations proceed safely with enforced verification between sessions.

Can I automate campaign continuation for cross-system feature builds that run overnight?

Yes, longer campaigns can be daemonized for automated continuation. The system creates git stash checkpoints and writes Continuation State, allowing cross-system feature builds to resume automatically while logging telemetry events for lifecycle tracking.

How do I enforce quality gates and verification steps before advancing to the next phase of a refactor?

Quality gates enforce verification by running typechecks, tests, regression guards, and anti-pattern scans before advancing. Phase end-condition checks ensure each stage of a long-running refactor meets explicit criteria before the campaign progresses to the next session.

What limitations should I consider when orchestrating sub-agents for multi-session campaigns?

Sub-agent orchestration depends on full context injection and lifecycle event logging to maintain consistency. If a campaign drifts, rollback procedures and Decision Logs support recovery, but complex migrations with inadequate phase decomposition may still suffer from inconsistent quality across delegated tasks.