gsd-cc 0.7.1 → 0.8.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (2) hide show
  1. package/package.json +1 -1
  2. package/skills/gsd/SKILL.md +36 -17
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "gsd-cc",
3
- "version": "0.7.1",
3
+ "version": "0.8.0",
4
4
  "description": "Get Shit Done on Claude Code — structured AI development with your Max plan",
5
5
  "author": "Philipp Briese (https://github.com/0ui-labs)",
6
6
  "homepage": "https://github.com/0ui-labs/GSD-CC#readme",
@@ -76,26 +76,45 @@ IF M*-ROADMAP.md exists AND there are slices without a S*-PLAN.md:
76
76
  ### Plan Ready, Not Executed
77
77
  ```
78
78
  IF S*-PLAN.md exists for current slice AND no T*-SUMMARY.md files for it:
79
- → Present the execution mode choice with a clear explanation:
79
+ → Present the three execution modes with clear pros/cons:
80
80
 
81
81
  "S{nn} is planned with {n} tasks. How do you want to execute?
82
82
 
83
- Manual mode:
84
- You work through each task one by one, each in a fresh session.
85
- After each task you can review the code, run tests, and decide
86
- whether to continue. More control, but you need to be present.
87
-
88
- Auto mode:
89
- Claude runs all tasks autonomously in the background via claude -p.
90
- Each task gets a fresh context window. UNIFY runs automatically
91
- when all tasks are done. You can walk away and come back later.
92
- Runs on your Max Plan no extra API costs.
93
- Downside: you can't intervene between tasks.
94
-
95
- manual or auto?"
96
-
97
- "manual" delegate to /gsd-cc-apply
98
- "auto" delegate to /gsd-cc-auto
83
+ 1) Manual
84
+ You work through each task one by one, each in a fresh session.
85
+ + Full control review code, run tests, adjust after each task
86
+ + You see exactly what happens
87
+ - You need to be present for every task
88
+ - Slowest option
89
+ Best for: critical slices, learning the codebase, first-time users
90
+
91
+ 2) Auto (this slice) ← recommended
92
+ Claude runs all {n} tasks in this slice autonomously.
93
+ UNIFY runs automatically when done.
94
+ Before the NEXT slice, you're back for Discuss + Plan.
95
+ + Tasks run in the background — go grab a coffee
96
+ + You still decide the direction for every slice
97
+ + Best balance of speed and control
98
+ - You can't intervene between tasks within this slice
99
+ Best for: most situations — you decide WHAT, Claude does the HOW
100
+
101
+ 3) Auto (full milestone)
102
+ Claude runs everything autonomously: plan, execute, UNIFY,
103
+ next slice, repeat — until the milestone is done.
104
+ Discuss is skipped. Claude makes all detail decisions.
105
+ + Fastest — walk away, come back when it's done
106
+ + Great for well-defined projects with tight rigor
107
+ - No input from you between slices
108
+ - Claude may make wrong assumptions in detail planning
109
+ - Higher risk of going in an unwanted direction
110
+ Best for: small/clear projects, utility tools, tight rigor
111
+
112
+ 1, 2, or 3?"
113
+
114
+ → "1" or "manual" → delegate to /gsd-cc-apply
115
+ → "2" or "auto" → delegate to /gsd-cc-auto (slice mode)
116
+ → "3" or "full auto" → delegate to /gsd-cc-auto (full milestone mode)
117
+ Set auto_mode_scope in STATE.md: "slice" or "milestone"
99
118
  ```
100
119
 
101
120
  ### Execution In Progress