get-shit-done-cc 1.4.6 → 1.4.8

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.
@@ -259,10 +259,10 @@ Use AskUserQuestion:
259
259
  - header: "Parallelization"
260
260
  - question: "Enable parallel phase execution?"
261
261
  - options:
262
- - "Disabled" — Execute plans sequentially (Recommended)
263
- - "Enabled" — Run independent plans in parallel (experimental, may not yield best results)
262
+ - "Enabled" — Run independent plans in parallel (Recommended)
263
+ - "Disabled" — Execute plans sequentially
264
264
 
265
- **Parallelization is experimental.** When enabled, `/gsd:execute-phase` spawns multiple agents for independent plans. Still being refined—sequential execution is more reliable. Can be changed later in config.json.
265
+ **Parallelization** spawns multiple agents for independent plans within a phase. Wave-based execution handles dependencies automatically. Can be changed later in config.json.
266
266
 
267
267
  </step>
268
268
 
@@ -6,8 +6,36 @@ Execute all plans in a phase using wave-based parallel execution. Orchestrator s
6
6
  The orchestrator's job is coordination, not execution. Each subagent loads the full execute-plan context itself. Orchestrator discovers plans, analyzes dependencies, groups into waves, spawns agents, handles checkpoints, collects results.
7
7
  </core_principle>
8
8
 
9
+ <required_reading>
10
+ Read STATE.md before any operation to load project context.
11
+ </required_reading>
12
+
9
13
  <process>
10
14
 
15
+ <step name="load_project_state" priority="first">
16
+ Before any operation, read project state:
17
+
18
+ ```bash
19
+ cat .planning/STATE.md 2>/dev/null
20
+ ```
21
+
22
+ **If file exists:** Parse and internalize:
23
+ - Current position (phase, plan, status)
24
+ - Accumulated decisions (constraints on this execution)
25
+ - Deferred issues (context for deviations)
26
+ - Blockers/concerns (things to watch for)
27
+
28
+ **If file missing but .planning/ exists:**
29
+ ```
30
+ STATE.md missing but planning artifacts exist.
31
+ Options:
32
+ 1. Reconstruct from existing artifacts
33
+ 2. Continue without project state (may lose accumulated context)
34
+ ```
35
+
36
+ **If .planning/ doesn't exist:** Error - project not initialized.
37
+ </step>
38
+
11
39
  <step name="validate_phase">
12
40
  Confirm phase exists and has plans:
13
41
 
@@ -277,6 +277,10 @@ Standard tasks (remain in standard plans):
277
277
  → Yes: Create a dedicated TDD plan for this feature (one feature per TDD plan)
278
278
  → No: Standard task in standard plan
279
279
 
280
+ **Why TDD gets its own plan:** TDD requires 2-3 execution cycles (RED → GREEN → REFACTOR), each with file reads, test runs, and potential debugging. Embedded in a multi-task plan, TDD work consumes 50-60% of context alone, degrading quality for remaining tasks.
281
+
282
+ **Test framework:** If project has no test setup and TDD plans are needed, the first TDD plan's RED phase handles framework setup as part of writing the first test.
283
+
280
284
  See `~/.claude/get-shit-done/references/tdd.md` for TDD plan structure.
281
285
 
282
286
  **Checkpoints:** Visual/functional verification → checkpoint:human-verify. Implementation choices → checkpoint:decision. Manual action (email, 2FA) → checkpoint:human-action (rare).
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "get-shit-done-cc",
3
- "version": "1.4.6",
3
+ "version": "1.4.8",
4
4
  "description": "A meta-prompting, context engineering and spec-driven development system for Claude Code by TÂCHES.",
5
5
  "bin": {
6
6
  "get-shit-done-cc": "bin/install.js"