gal-Tab

Self-improving skills loop. Reads recent sessions (index + axcli analytics + transcripts), surfaces recommended skill edits — new trigger phrases, description tweaks, skill merges, new-skill candidates — and emits them as a draft for Gal to approve. Never…

Decision memory: record a call once, auto-apply it every time the same question recurs — never re-decide what's already settled. Use when the user says "/decide", "record this decision", "we decided", "did we decide X before", "log this call", or "is this…

The Discover/Think front door — a Core phase that picks the right thinking lens BEFORE a GTM, Building, or Automation task starts. Front-runs all three working modes; load it before committing to any approach or tool. Triggers: "help me think through X",…

Scope Edit/Write to a directory during debugging or investigation so unrelated files are not accidentally modified. Use when the user says "/freeze", "/freeze <path>", "freeze this path", "scope edits to", "lock to this dir", "/unfreeze", "freeze off".

Close the compound-learning loop: surface past corrections/patterns/playbooks from the project's .compound/ directory, then canonicalize accumulated drafts before they reach wiki-compile. Use when the user says "learnings", "load past corrections", "what did…

HubSpot API-safety check. Before shipping a workflow or code that calls HubSpot, scan for deprecated v1 endpoints and flag them with current v3 replacements so nothing silently breaks in production. Triggers: "hubspot safety", "check hubspot api", "is this…

Triage a failed n8n workflow execution. Paste an n8n execution URL and this skill pulls the failed node, reads the error, and proposes a concrete fix. Uses the n8n-mcp server (n8n at n8n.bigbrain.me). Triggers: "/n8n-triage", "n8n execution failed", "debug…

Defines the shared Building loop and named sub-loop entry points so dashboards, wikis, and MCPs each follow a consistent but distinct flow. Use when the user says "build discipline", "how do I build X", "start a new dashboard", "start a new wiki", "start a…

Use when the user asks what was learned before, references past mistakes/lessons/playbooks ("have we hit this before", "what did we learn about X", "lessons learned", "did I already solve this"), or when you want to consult the agent's own compound learnings…

Use when planning or brainstorming a non-trivial task and you want a thorough, one-shot sweep of the agent's past learnings before committing to an approach — best run as a dispatched subagent so the scan happens in its own token budget and returns only a…

Use when user asks domain questions and a wiki/ directory exists in the project, when user references "my research" or "what I've read," when user asks about topics that could be in their knowledge base, or when user asks to compare entities from ingested…