Shadow Evolution Assimilation

🧬 Shadow Evolution Protocol

This skill empowers the Shadow agent to proactively "hunt" for improvements in child projects and assimilate them into the current Mother Hive template.

🛡️ Prerequisites

• Must be running within the shadow-genesis (Mother Hive) context.
• A valid child project path must be provided (or scan all from child_registry.json).

📜 Workflow

1. Preparation (Backup)

CRITICAL: Before any assimilation involves modifying the Mother Hive, create a backup via Shadow Chronos.

2. Genetic Scanning (Smart Diff)

PRIMARY SOURCE: Read the child's mutation_log.json first!

• Location: <ChildNode>/.shadow/memory/life/resources/system/mutation_log.json
• Filter by Status: Only show mutations with status: "pending_review".

FALLBACK (if no mutation_log exists): Use blind scan:

• Filters:
  • IGNORE: Project-specific memory (memory/life/projects/*), logs (memory/*.md), user-specific tweaks.
  • TARGET:
    • New Skills (.shadow/skills/*)
    • Improved System Prompts (tacit_knowledge.md)
    • New Workflows (.agent/workflows/*)
    • Improved Scripts

3. Assimilation (Merge)

• Present the found "mutations" (changes) to the Master.

• For each approved mutation:

  1. Copy the file/folder from Child to Mother.
  2. Update the child's mutation_log.json: Set status to "assimilated".
  3. Append a record to Mother's assimilation_history.json.

• For each rejected mutation:

  1. Update the child's mutation_log.json: Set status to "rejected".

4. Post-Assimilation

• Prompt for Snapshot: Per the "Evolution Law", always ask if a new version snapshot should be created.
• Update version_history.json if snapshot is approved.

💻 Usage

User: "Evolution"
Agent: [Scans all children's mutation_log.json, presents pending mutations]

User: "Assimilate the new skill from InduLog."
Agent: [Executes Skill: Shadow Evolution] --Target "InduLog" --MutationId "mut-001"

Converted and distributed by TomeVault — claim your Tome and manage your conversions.