synapse

Diagnose SYNAPSE context rule injection failures across agent, workflow, and task states.

Updated May 28, 2026
One-click install
npx skills add https://github.com/GustaPaes/aiox-sprint-canvas --skill synapse-gustapaes
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: synapse
Source: https://github.com/GustaPaes/aiox-sprint-canvas/tree/main/.claude/skills/synapse
Command: npx skills add https://github.com/GustaPaes/aiox-sprint-canvas --skill synapse-gustapaes

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) and assets (resource) components.

What problem does it solve?

SYNAPSE helps you understand and manage how AIOX injects the right context rules into each prompt, so answers stay consistent with your active domains, workflow, task, and mode.

Core Features & Use Cases

  • Explain SYNAPSE context behavior: Understand the 8-layer processing pipeline and how <synapse-rules> is produced for each prompt.
  • Manage context domains: List domains, create new domains, and troubleshoot why a domain is or isn’t loading.
  • Configure injection via brackets: Use the FRESH/MODERATE/DEPLETED/CRITICAL bracket system to predict and reason about how much context gets injected.
  • Debug rule injection: Inspect loaded layers, rule counts, and manifest/domain parsing outcomes using debug/status commands.
  • Use star-commands and modes: Switch response behavior (e.g., *brief, *dev, *review) and see how it maps into injected rules.

Example use case: When your agent changes from dev to QA during a workflow, SYNAPSE ensures the prompt receives the correct agent-scoped and workflow-scoped rule sets.

Quick Start

Ask: "What domains are currently active, what context bracket am I in, and why are the injected rules missing or incorrect?"

Frequently Asked Questions about synapse

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

FAQPage Schema
How does context injection work for agent prompts?

Domain management controls which scoped rule sets load into the prompt. You can list active domains, create new ones, and troubleshoot loading failures using CRUD operations on domains and manifests to verify correct layer processing.

How do I debug missing or incorrect context rules?

Bracket-based layer activation uses FRESH, MODERATE, DEPLETED, and CRITICAL states to control how much context gets injected into each prompt. This system helps predict and reason about injected context volume across different operational states.

Can I switch agent response behavior during a workflow?

Star-commands switch response behavior by mapping modes like *brief, *dev, and *review into injected rules. When an agent changes roles during a workflow, the system ensures the prompt receives the correct agent-scoped and workflow-scoped rule sets.

Why are my injected context rules missing or incorrect?

Injected context rules fail due to incorrect domain loading, manifest parsing errors, or improper bracket layer activation. Diagnose these injection failures by inspecting loaded layers and rule counts with debug commands to ensure correct formatting in the synapse-rules block.