~cytrogen/gstack

9c5f479745acc90533a7ff75a00771b9056c43ef — Cytrogen 2 days ago cf73db5 main
fork: 频率分级路由 + 触发式描述符重写

- 4 个高频 skill (browse/investigate/qa/office-hours) 重写描述为触发式
- 32 个低频 skill 加 disable-model-invocation: true
- .agents/ 和 .factory/ 内副本同步禁用自动加载
- gstack 主入口 SKILL.md 禁用自动加载(与 browse 重复)
M SKILL.md => SKILL.md +1 -0
@@ 1,5 1,6 @@
---
name: gstack
disable-model-invocation: true
preamble-tier: 1
version: 1.1.0
description: |

M autoplan/SKILL.md => autoplan/SKILL.md +1 -0
@@ 1,5 1,6 @@
---
name: autoplan
disable-model-invocation: true
preamble-tier: 3
version: 1.0.0
description: |

M benchmark/SKILL.md => benchmark/SKILL.md +1 -0
@@ 1,5 1,6 @@
---
name: benchmark
disable-model-invocation: true
preamble-tier: 1
version: 1.0.0
description: |

M bin/gstack-global-discover => bin/gstack-global-discover +0 -0
M browse/SKILL.md => browse/SKILL.md +2 -7
@@ 1,14 1,9 @@
---
name: browse
disable-model-invocation: true
preamble-tier: 1
version: 1.1.0
description: |
  Fast headless browser for QA testing and site dogfooding. Navigate any URL, interact with
  elements, verify page state, diff before/after actions, take annotated screenshots, check
  responsive layouts, test forms and uploads, handle dialogs, and assert element states.
  ~100ms per command. Use when you need to test a feature, verify a deployment, dogfood a
  user flow, or file a bug with evidence. Use when asked to "open in browser", "test the
  site", "take a screenshot", or "dogfood this". (gstack)
description: Invoke when you need to open a URL, test a web page, take screenshots, verify a deployment, or dogfood a user flow. Use for all web browsing needs. Not for reading docs as text or fetching raw content.
allowed-tools:
  - Bash
  - Read

M canary/SKILL.md => canary/SKILL.md +1 -0
@@ 1,5 1,6 @@
---
name: canary
disable-model-invocation: true
preamble-tier: 2
version: 1.0.0
description: |

M careful/SKILL.md => careful/SKILL.md +1 -0
@@ 1,5 1,6 @@
---
name: careful
disable-model-invocation: true
version: 0.1.0
description: |
  Safety guardrails for destructive commands. Warns before rm -rf, DROP TABLE,

M checkpoint/SKILL.md => checkpoint/SKILL.md +1 -0
@@ 1,5 1,6 @@
---
name: checkpoint
disable-model-invocation: true
preamble-tier: 2
version: 1.0.0
description: |

M codex/SKILL.md => codex/SKILL.md +1 -0
@@ 1,5 1,6 @@
---
name: codex
disable-model-invocation: true
preamble-tier: 3
version: 1.0.0
description: |

M connect-chrome/SKILL.md => connect-chrome/SKILL.md +1 -0
@@ 1,5 1,6 @@
---
name: connect-chrome
disable-model-invocation: true
version: 0.1.0
description: |
  Launch real Chrome controlled by gstack with the Side Panel extension auto-loaded.

M cso/SKILL.md => cso/SKILL.md +1 -0
@@ 1,5 1,6 @@
---
name: cso
disable-model-invocation: true
preamble-tier: 2
version: 2.0.0
description: |

M design-consultation/SKILL.md => design-consultation/SKILL.md +1 -0
@@ 1,5 1,6 @@
---
name: design-consultation
disable-model-invocation: true
preamble-tier: 3
version: 1.0.0
description: |

M design-html/SKILL.md => design-html/SKILL.md +1 -0
@@ 1,5 1,6 @@
---
name: design-html
disable-model-invocation: true
preamble-tier: 2
version: 1.0.0
description: |

M design-review/SKILL.md => design-review/SKILL.md +1 -0
@@ 1,5 1,6 @@
---
name: design-review
disable-model-invocation: true
preamble-tier: 4
version: 2.0.0
description: |

M design-shotgun/SKILL.md => design-shotgun/SKILL.md +1 -0
@@ 1,5 1,6 @@
---
name: design-shotgun
disable-model-invocation: true
preamble-tier: 2
version: 1.0.0
description: |

M devex-review/SKILL.md => devex-review/SKILL.md +1 -0
@@ 1,5 1,6 @@
---
name: devex-review
disable-model-invocation: true
preamble-tier: 3
version: 1.0.0
description: |

M document-release/SKILL.md => document-release/SKILL.md +1 -0
@@ 1,5 1,6 @@
---
name: document-release
disable-model-invocation: true
preamble-tier: 2
version: 1.0.0
description: |

M freeze/SKILL.md => freeze/SKILL.md +1 -0
@@ 1,5 1,6 @@
---
name: freeze
disable-model-invocation: true
version: 0.1.0
description: |
  Restrict file edits to a specific directory for the session. Blocks Edit and

M gstack-upgrade/SKILL.md => gstack-upgrade/SKILL.md +1 -0
@@ 1,5 1,6 @@
---
name: gstack-upgrade
disable-model-invocation: true
version: 1.1.0
description: |
  Upgrade gstack to the latest version. Detects global vs vendored install,

M guard/SKILL.md => guard/SKILL.md +1 -0
@@ 1,5 1,6 @@
---
name: guard
disable-model-invocation: true
version: 0.1.0
description: |
  Full safety mode: destructive command warnings + directory-scoped edits.

M health/SKILL.md => health/SKILL.md +1 -0
@@ 1,5 1,6 @@
---
name: health
disable-model-invocation: true
preamble-tier: 2
version: 1.0.0
description: |

M investigate/SKILL.md => investigate/SKILL.md +2 -8
@@ 1,15 1,9 @@
---
name: investigate
disable-model-invocation: true
preamble-tier: 2
version: 1.0.0
description: |
  Systematic debugging with root cause investigation. Four phases: investigate,
  analyze, hypothesize, implement. Iron Law: no fixes without root cause.
  Use when asked to "debug this", "fix this bug", "why is this broken",
  "investigate this error", or "root cause analysis".
  Proactively invoke this skill (do NOT debug directly) when the user reports
  errors, 500 errors, stack traces, unexpected behavior, "it was working
  yesterday", or is troubleshooting why something stopped working. (gstack)
description: Invoke when the user reports a bug, error, stack trace, or unexpected behavior. Must use this instead of debugging directly. Not for feature requests, code review, or small known fixes.
allowed-tools:
  - Bash
  - Read

M land-and-deploy/SKILL.md => land-and-deploy/SKILL.md +1 -0
@@ 1,5 1,6 @@
---
name: land-and-deploy
disable-model-invocation: true
preamble-tier: 4
version: 1.0.0
description: |

M learn/SKILL.md => learn/SKILL.md +1 -0
@@ 1,5 1,6 @@
---
name: learn
disable-model-invocation: true
preamble-tier: 2
version: 1.0.0
description: |

M office-hours/SKILL.md => office-hours/SKILL.md +2 -12
@@ 1,19 1,9 @@
---
name: office-hours
disable-model-invocation: true
preamble-tier: 3
version: 2.0.0
description: |
  YC Office Hours — two modes. Startup mode: six forcing questions that expose
  demand reality, status quo, desperate specificity, narrowest wedge, observation,
  and future-fit. Builder mode: design thinking brainstorming for side projects,
  hackathons, learning, and open source. Saves a design doc.
  Use when asked to "brainstorm this", "I have an idea", "help me think through
  this", "office hours", or "is this worth building".
  Proactively invoke this skill (do NOT answer directly) when the user describes
  a new product idea, asks whether something is worth building, wants to think
  through design decisions for something that doesn't exist yet, or is exploring
  a concept before any code is written.
  Use before /plan-ceo-review or /plan-eng-review. (gstack)
description: Invoke when the user has a new idea, wants to brainstorm, or asks "is this worth building". Use before any planning skills. Not for existing bugs, implementation tasks, or code architecture decisions.
allowed-tools:
  - Bash
  - Read

M plan-ceo-review/SKILL.md => plan-ceo-review/SKILL.md +1 -0
@@ 1,5 1,6 @@
---
name: plan-ceo-review
disable-model-invocation: true
preamble-tier: 3
version: 1.0.0
description: |

M plan-design-review/SKILL.md => plan-design-review/SKILL.md +1 -0
@@ 1,5 1,6 @@
---
name: plan-design-review
disable-model-invocation: true
preamble-tier: 3
version: 2.0.0
description: |

M plan-devex-review/SKILL.md => plan-devex-review/SKILL.md +1 -0
@@ 1,5 1,6 @@
---
name: plan-devex-review
disable-model-invocation: true
preamble-tier: 3
version: 1.0.0
description: |

M plan-eng-review/SKILL.md => plan-eng-review/SKILL.md +1 -0
@@ 1,5 1,6 @@
---
name: plan-eng-review
disable-model-invocation: true
preamble-tier: 3
version: 1.0.0
description: |

M qa-only/SKILL.md => qa-only/SKILL.md +1 -0
@@ 1,5 1,6 @@
---
name: qa-only
disable-model-invocation: true
preamble-tier: 4
version: 1.0.0
description: |

M qa/SKILL.md => qa/SKILL.md +2 -10
@@ 1,17 1,9 @@
---
name: qa
disable-model-invocation: true
preamble-tier: 4
version: 2.0.0
description: |
  Systematically QA test a web application and fix bugs found. Runs QA testing,
  then iteratively fixes bugs in source code, committing each fix atomically and
  re-verifying. Use when asked to "qa", "QA", "test this site", "find bugs",
  "test and fix", or "fix what's broken".
  Proactively suggest when the user says a feature is ready for testing
  or asks "does this work?". Three tiers: Quick (critical/high only),
  Standard (+ medium), Exhaustive (+ cosmetic). Produces before/after health scores,
  fix evidence, and a ship-readiness summary. For report-only mode, use /qa-only. (gstack)
  Voice triggers (speech-to-text aliases): "quality check", "test the app", "run QA".
description: Invoke when a feature is ready for testing, the user asks to find bugs in a running site, or says "does this work". Tests and fixes bugs. For report-only without fixes, use /qa-only instead. Not for unit tests or code review.
allowed-tools:
  - Bash
  - Read

M retro/SKILL.md => retro/SKILL.md +1 -0
@@ 1,5 1,6 @@
---
name: retro
disable-model-invocation: true
preamble-tier: 2
version: 2.0.0
description: |

M review/SKILL.md => review/SKILL.md +1 -0
@@ 1,5 1,6 @@
---
name: review
disable-model-invocation: true
preamble-tier: 4
version: 1.0.0
description: |

M setup-browser-cookies/SKILL.md => setup-browser-cookies/SKILL.md +1 -0
@@ 1,5 1,6 @@
---
name: setup-browser-cookies
disable-model-invocation: true
preamble-tier: 1
version: 1.0.0
description: |

M setup-deploy/SKILL.md => setup-deploy/SKILL.md +1 -0
@@ 1,5 1,6 @@
---
name: setup-deploy
disable-model-invocation: true
preamble-tier: 2
version: 1.0.0
description: |

M ship/SKILL.md => ship/SKILL.md +1 -0
@@ 1,5 1,6 @@
---
name: ship
disable-model-invocation: true
preamble-tier: 4
version: 1.0.0
description: |

M unfreeze/SKILL.md => unfreeze/SKILL.md +1 -0
@@ 1,5 1,6 @@
---
name: unfreeze
disable-model-invocation: true
version: 0.1.0
description: |
  Clear the freeze boundary set by /freeze, allowing edits to all directories