@mthanhlm/autodev 0.3.5 → 0.3.6

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.
@@ -1,7 +1,7 @@
1
1
  {
2
2
  "name": "autodev",
3
3
  "description": "A lean Claude Code workflow system with a single entrypoint, task-based phase execution, and read-only git.",
4
- "version": "0.3.5",
4
+ "version": "0.3.6",
5
5
  "author": {
6
6
  "name": "mthanhlm"
7
7
  },
package/README.md CHANGED
@@ -89,7 +89,7 @@ project -> track -> phase -> tasks
89
89
 
90
90
  - Large phases are too big for one uninterrupted execution context
91
91
  - Developers need control over what changes next
92
- - Each task can run in a fresh background agent context
92
+ - Each task can run in a fresh foreground agent context
93
93
  - The main `/autodev` session stays small and orchestration-focused
94
94
 
95
95
  ### Execution Model
@@ -98,7 +98,7 @@ project -> track -> phase -> tasks
98
98
  - A phase is planned into one phase overview plus multiple task files
99
99
  - `/autodev` stops after planning so the user can review the phase before any execution starts
100
100
  - The phase keeps one user-facing orchestration session
101
- - Each task is preferably executed by a fresh worker agent in the background
101
+ - Each task is preferably executed by a fresh foreground delegated agent
102
102
  - After each task, the worker reports back with files changed, verification, and blockers
103
103
  - No waves by default
104
104
  - No parallel execution unless the user explicitly asks for it later
@@ -2,6 +2,8 @@
2
2
  name: autodev-codebase-domain
3
3
  description: Product and domain mapper for brownfield exploration. Use proactively when mapping an existing codebase and writing `.autodev/codebase/domain.md`.
4
4
  tools: Read, Grep, Glob, Bash, Write
5
+ background: false
6
+ color: purple
5
7
  model: inherit
6
8
  ---
7
9
 
@@ -2,6 +2,8 @@
2
2
  name: autodev-codebase-quality
3
3
  description: Testing and technical-debt mapper for brownfield exploration. Use proactively when mapping an existing codebase and writing `.autodev/codebase/quality.md`.
4
4
  tools: Read, Grep, Glob, Bash, Write
5
+ background: false
6
+ color: yellow
5
7
  model: inherit
6
8
  ---
7
9
 
@@ -2,6 +2,8 @@
2
2
  name: autodev-codebase-runtime
3
3
  description: Runtime and dependency mapper for brownfield exploration. Use proactively when mapping an existing codebase and writing `.autodev/codebase/runtime.md`.
4
4
  tools: Read, Grep, Glob, Bash, Write
5
+ background: false
6
+ color: blue
5
7
  model: inherit
6
8
  ---
7
9
 
@@ -2,6 +2,8 @@
2
2
  name: autodev-codebase-structure
3
3
  description: Repository structure mapper for brownfield exploration. Use proactively when mapping an existing codebase and writing `.autodev/codebase/structure.md`.
4
4
  tools: Read, Grep, Glob, Bash, Write
5
+ background: false
6
+ color: cyan
5
7
  model: inherit
6
8
  ---
7
9
 
@@ -2,6 +2,8 @@
2
2
  name: autodev-review-integration
3
3
  description: Integration review specialist for executed phases. Use proactively after code changes to find cross-module regressions and wiring mistakes.
4
4
  tools: Read, Grep, Glob, Bash
5
+ background: false
6
+ color: orange
5
7
  model: inherit
6
8
  ---
7
9
 
@@ -2,6 +2,8 @@
2
2
  name: autodev-review-polish
3
3
  description: Product polish review specialist for executed phases. Use proactively after code changes to find rough UX, copy, and finish issues.
4
4
  tools: Read, Grep, Glob, Bash
5
+ background: false
6
+ color: pink
5
7
  model: inherit
6
8
  ---
7
9
 
@@ -2,6 +2,8 @@
2
2
  name: autodev-review-quality
3
3
  description: Quality review specialist for executed phases. Use proactively after code changes to find correctness, maintainability, and test issues.
4
4
  tools: Read, Grep, Glob, Bash
5
+ background: false
6
+ color: green
5
7
  model: inherit
6
8
  ---
7
9
 
@@ -2,6 +2,8 @@
2
2
  name: autodev-review-security
3
3
  description: Security review specialist for executed phases. Use proactively after code changes to find auth, input-handling, and trust-boundary issues.
4
4
  tools: Read, Grep, Glob, Bash
5
+ background: false
6
+ color: red
5
7
  model: inherit
6
8
  ---
7
9
 
@@ -1,7 +1,9 @@
1
1
  ---
2
2
  name: autodev-task-worker
3
- description: Single-task implementation worker for autodev phases. Use proactively when one `.autodev` task should run in a fresh agent context and produce its matching `TASK-NN-SUMMARY.md`.
3
+ description: Single-task implementation specialist for autodev phases. Use proactively when one `.autodev` task should run in a fresh foreground agent context and produce its matching `TASK-NN-SUMMARY.md`.
4
4
  tools: Read, Grep, Glob, Bash, Edit, Write
5
+ background: false
6
+ color: orange
5
7
  model: inherit
6
8
  ---
7
9
 
@@ -33,7 +33,7 @@ Type: [feature|bugfix|refactor|research|polish]
33
33
  | 02 | pending | 01 | [Concrete task goal] |
34
34
 
35
35
  ## Notes
36
- - [Anything the worker agents must know]
36
+ - [Anything the delegated agents must know]
37
37
 
38
38
  ## Git Policy
39
39
  - Read-only only.
@@ -26,7 +26,7 @@ node "$AUTODEV_ROOT/autodev/bin/autodev-tools.cjs" status
26
26
  - `explore_codebase`: load and perform the explore-codebase workflow.
27
27
  - `plan_phase`: load and perform the plan-phase workflow.
28
28
  - `plan_review`: load and perform the review-plan workflow so the user can review the phase plan before any execution starts.
29
- - `execute_phase`: load and perform the execute-phase workflow, which should advance one task-sized unit in the current phase session by trying a fresh worker agent first and falling back cleanly when agent delegation is unavailable.
29
+ - `execute_phase`: load and perform the execute-phase workflow, which should advance one task-sized unit in the current phase session by trying a fresh foreground agent first and falling back cleanly when agent delegation is unavailable.
30
30
  - `review_phase`: load and perform the review-phase workflow.
31
31
  - `verify_phase`: load and perform the verify-work workflow.
32
32
 
@@ -1,5 +1,5 @@
1
1
  <purpose>
2
- Execute an active-track phase through task-sized worker runs, keep the main session orchestration-focused, and capture both task summaries and the final phase summary.
2
+ Execute an active-track phase through task-sized delegated runs, keep the main session orchestration-focused, and capture both task summaries and the final phase summary.
3
3
  </purpose>
4
4
 
5
5
  <rules>
@@ -8,11 +8,12 @@ Execute an active-track phase through task-sized worker runs, keep the main sess
8
8
  - Stay within the current phase. Do not silently expand scope.
9
9
  - Execute one task at a time by default.
10
10
  - Do not use waves.
11
- - Prefer a fresh background worker for each task so the phase session does not bloat.
11
+ - Prefer a fresh foreground agent for each task so the phase session does not bloat.
12
12
  - If agent delegation is available in this environment, delegate task implementation.
13
13
  - If a delegation attempt fails with an environment limitation like `specialized agents aren't available`, treat agent delegation as unavailable for this run and continue cleanly.
14
14
  - The main phase session should remain orchestration-focused whenever delegation works.
15
15
  - The main phase session may update `.autodev/` state and aggregate phase artifacts.
16
+ - Do not run autodev subagents as background tasks.
16
17
  </rules>
17
18
 
18
19
  <process>
@@ -54,6 +55,8 @@ node "$AUTODEV_ROOT/autodev/bin/autodev-tools.cjs" init execute-phase "$ARGUMENT
54
55
  - `autodev-task-worker` when autodev is installed into `.claude/`
55
56
  - `autodev:autodev-task-worker` only when this workflow is being run directly from a plugin directory
56
57
 
58
+ Run it in the foreground as part of the current workflow step.
59
+
57
60
  Tell the subagent:
58
61
  - phase path
59
62
  - task path
@@ -70,7 +73,7 @@ node "$AUTODEV_ROOT/autodev/bin/autodev-tools.cjs" init execute-phase "$ARGUMENT
70
73
  - keep the same task boundaries
71
74
  - still write the task summary before moving on
72
75
 
73
- 9. After the worker returns, or after the current-session fallback completes:
76
+ 9. After the delegated agent returns, or after the current-session fallback completes:
74
77
  - confirm `TASK-NN-SUMMARY.md` exists
75
78
  - inspect whether the task is done or blocked
76
79
  - report the result to the user
@@ -18,7 +18,7 @@ Lean Claude Code workflow. No automatic commits. No branches. No worktrees. Git
18
18
  - `/autodev-plan-phase [phase]`
19
19
  Creates or revises one phase plan plus task files in `.autodev/tracks/<track>/phases/NN-type-name/`.
20
20
  - `/autodev-execute-phase [phase]`
21
- Orchestrates one phase task-by-task, preferring a fresh worker agent per task, and writes `TASK-NN-SUMMARY.md` plus the final `NN-SUMMARY.md`.
21
+ Orchestrates one phase task-by-task, preferring a fresh foreground delegated agent per task, and writes `TASK-NN-SUMMARY.md` plus the final `NN-SUMMARY.md`.
22
22
  - `/autodev-review-phase [phase]`
23
23
  Uses four review agents when available for code quality, security, integration, and polish, then writes `NN-REVIEW.md`.
24
24
  - `/autodev-verify-work [phase]`
@@ -1,5 +1,5 @@
1
1
  <purpose>
2
- Create one practical phase plan for the active track, then break it into reviewable task files that are small enough for fresh worker-agent execution.
2
+ Create one practical phase plan for the active track, then break it into reviewable task files that are small enough for fresh foreground-agent execution.
3
3
  </purpose>
4
4
 
5
5
  <rules>
@@ -58,7 +58,7 @@ TASK-02.md
58
58
  ```
59
59
 
60
60
  Each task should be:
61
- - small enough for one fresh worker agent context
61
+ - small enough for one fresh delegated agent context
62
62
  - clear about likely files to touch
63
63
  - explicit about verification
64
64
  - reviewable as a standalone unit
@@ -1,6 +1,6 @@
1
1
  ---
2
2
  name: autodev:execute-phase
3
- description: Execute an active-track phase task by task, preferring fresh background workers with clean fallback when unavailable
3
+ description: Execute an active-track phase task by task, preferring fresh foreground agents with clean fallback when unavailable
4
4
  argument-hint: "[phase-number]"
5
5
  allowed-tools:
6
6
  - Read
@@ -14,7 +14,7 @@ allowed-tools:
14
14
  - Agent
15
15
  ---
16
16
  <objective>
17
- Execute one active-track phase through task-sized worker runs and record both task-level and phase-level summaries.
17
+ Execute one active-track phase through task-sized delegated runs and record both task-level and phase-level summaries.
18
18
  </objective>
19
19
 
20
20
  <execution_context>
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@mthanhlm/autodev",
3
- "version": "0.3.5",
3
+ "version": "0.3.6",
4
4
  "description": "A lean Claude Code workflow system with a single entrypoint, task-based phase execution, and read-only git.",
5
5
  "bin": {
6
6
  "autodev": "bin/install.js"