Name: discover
Author: gal-Tab

Imported from gal-Tab/sir_albert (plugins/sir-albert/skills/core/discover/SKILL.md). Install upstream with npx skills add gal-Tab/sir_albert --skill discover. Copyright stays with the author.

Discover

Boot from os/PREAMBLE.md.

Purpose

Discover is the front door to thinking — a Core phase (mode-agnostic) that runs before committing to a GTM, Building, or Automation task. Its job is to pick the right lens, not to execute. Getting this wrong costs more than the time it takes to do it right.

The empirical case: "WDYT?" ×43 and option-picking moments ×77 across sessions show this phase is constantly happening but inconsistently routed. This skill names it and routes it.

Routing table

Read the situation, pick one route. Offer the match and ask to confirm before loading the skill.

Situation	Route to
Explore an idea / fresh AI-first perspective	`discovery-lens`
Want multiple independent viewpoints	`board-of-advisors`
Attack / challenge / find holes	`devils-advocate`
Stress-test my thinking (interview me)	`grilling` — or `grill-with-docs` when there are docs to sharpen against
Bigger picture / how does this fit	`zoom-out`
Pin down terminology / record an ADR	`domain-modeling`
Sanity-check a UI or state model quickly	`prototype`

If the situation matches more than one row, surface the top two and ask which matters most right now.

Discover is a phase, not a pack

It precedes execute/build tasks — it does not replace them. Once a lens is chosen and the thinking has produced a clear direction, Discover is done. Hand off to the appropriate working skill (e.g. to-prd, domain-modeling, prototype, or a biz/dev/agentic skill).

Do not linger here. The goal is a sharper starting point for the next thing, not a permanent mode.

How to run

Ask one clarifying question if the situation is ambiguous: "Are you trying to explore this fresh, stress-test a formed view, or just figure out where to start?"
Name the matched lens and why.
Load the skill and hand off.

Keep this interaction tight — two or three turns maximum before routing.

{ "ocm": "1", "id": "gal-tab-sir-albert-discover", "kind": "skill", "name": "discover", "description": "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\", \"let's explore\", \"discover\", \"think this through\", \"which lens\", \"office hours\", \"WDYT\", \"where do I start with\", \"not sure how to approach\", or any moment where choosing the right lens matters more than executing fast.", "publisher": "gal-Tab", "version": "1.0.0", "capabilities": { "domains": [ "data_analysis" ], "tags": [ "skill-md", "github" ], "languages": [ "en" ] }, "quality_prior": 0.6, "examples": [ "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\", \"let's explore\", \"discover\", \"think this through\", \"which lens\", \"office hours\", \"WDYT\", \"where do I start with\", \"not sure how to approach\", or any moment where choosing the right lens matters more than executing fast." ], "primary": false, "metadata": { "source": { "provider": "github", "repository": "https://github.com/gal-Tab/sir_albert", "path": "plugins/sir-albert/skills/core/discover/SKILL.md", "ref": "94a4188c496674a69a7b6fa4c2317e1cf6f90150", "url": "https://github.com/gal-Tab/sir_albert/blob/94a4188c496674a69a7b6fa4c2317e1cf6f90150/plugins/sir-albert/skills/core/discover/SKILL.md", "key": "gal-Tab/sir_albert/plugins/sir-albert/skills/core/discover/SKILL.md" } }, "instructions": "# Discover\n\nBoot from `os/PREAMBLE.md`.\n\n## Purpose\n\nDiscover is the **front door to thinking** — a Core phase (mode-agnostic) that runs before committing to a GTM, Building, or Automation task. Its job is to pick the right lens, not to execute. Getting this wrong costs more than the time it takes to do it right.\n\nThe empirical case: \"WDYT?\" ×43 and option-picking moments ×77 across sessions show this phase is constantly happening but inconsistently routed. This skill names it and routes it.\n\n---\n\n## Routing table\n\nRead the situation, pick one route. Offer the match and ask to confirm before l", "cost": { "context_tokens": 492 } }

discover

About

Discover

Purpose

Routing table

Discover is a phase, not a pack

How to run

Use it

Manifest

Reviews