Claude Skills by friday-ai-codes

Create a clean PR branch by filtering out .planning/ commits — ready for code review

Generate developer behavioral profile and create Claude-discoverable artifacts

Check progress, advance workflow, or dispatch freeform intent — the unified GSD situational command

Execute a quick task with GSD guarantees (atomic commits, state tracking) but skip optional agents

Resume work from previous session with full context restoration

Review and promote backlog items to active milestone

Request cross-AI peer review of phase plans from external AI CLIs

Retroactively verify threat mitigations for a completed phase

Configure GSD workflow toggles and model profile

Create PR, run review, and prepare for merge after verification passes

Sketch UI/design ideas with throwaway HTML mockups, or propose what to sketch next (frontier mode)

Clarify WHAT a phase delivers with ambiguity scoring; produces a SPEC.md before discuss-phase.

Spike an idea through experiential exploration, or propose what to spike next (frontier mode)

Display project statistics — phases, plans, requirements, git metrics, and timeline

Toggle which skills are surfaced — apply a profile, list, or disable a cluster without reinstall

Manage persistent context threads for cross-session work

Generate UI design contract (UI-SPEC.md) for frontend phases

Retroactive 6-pillar visual audit of implemented frontend code

[BETA] Offload plan phase to Cursor's ultraplan cloud; review in browser and import back.

Safe git revert. Roll back phase or plan commits using the phase manifest with dependency checks.

Update GSD to latest version with changelog display

Retroactively audit and fill Nyquist validation gaps for a completed phase

Validate built features through conversational UAT

Manage GSD workspaces — create, list, or remove isolated workspace environments

Manage parallel workstreams — list, create, switch, status, progress, complete, and resume

当用户要 Friday 操作或追踪某个远端已索引仓库时使用：找仓库（哪个仓库负责 X、某接口的后端/前端在哪个仓库）、跨仓库追接口实现（前端接口→后端实现、后端接口→前端调用点、检索命中请求 URL/路由定义/接口调用）、分析架构/风险/影响面（X 在仓库 Y 里怎么实现的）、生成或修改编码计划、执行计划并创建 MR，或给一个具体需求一条龙跑到 MR。覆盖发现→分析→计划→执行全流程。反向边界：目标就在当前本地工作区、且不涉及另一侧仓库时用本地工具。

当用户要做 impact-analysis / 影响面分析时使用：改这段代码影响哪些调用方与执行流、MR/PR 风险面、detect_changes 看工作区改动波及范围、list_processes 看受影响执行流。覆盖 context/staleness 判读 → detect_changes/impact/list_processes → 解读 affected_processes。反向边界：只要仓库路由落点用 friday-routing；只要完整编码到 MR 用 friday-code；符号改名清单用 friday-refactoring。

Friday 的记忆层。当用户要把完成的任务沉淀为可复用经验（LearningCase）、在新工作前检索相似需求/缺陷修复经验，或查询历史交付知识——「以前做过类似需求吗」、实体版本时间线、需求→方案→MR 关联链、历史时点 as_of 查询——时使用。反向边界：当前任务无需历史经验/交付检索时，不必走本技能。

当用户要做 refactoring / 符号改名 / 重命名影响清单时使用：先 rename_preview 拿只读双源编辑清单（图引用 + 文本兜底），再自行编辑；不申请 apply。覆盖触发条件、preview→编辑顺序与失败不阻断交付。反向边界：只要影响面/执行流分析用 friday-impact；完整需求到 MR 用 friday-code。