expand-tasks

Name: expand-tasks
Author: anombyte93

582stars57forks

How to install expand-tasks?

anombyte93/prd-taskmaster/expand-tasks

$npx -y skills add anombyte93/prd-taskmaster --skill expand-tasks

Installs into the current project.

Summary

Expand all TaskMaster tasks with deep research before coding begins. Reads tasks.json, launches parallel research agents per task in waves using the research-expander agent. Writes findings back to tasks.json. Part of the prd-taskmaster toolkit. Use after PRD is parsed and before

Files · 1

View on GitHub

SKILL.md

1	# Expand Tasks with Research v1.0
2
3	Expands TaskMaster tasks with research before coding begins.
4	Deterministic operations handled by `script.py`; AI handles judgment.
5
6	Script location: `skills/expand-tasks/script.py` (relative to plugin root)
7	Part of: `prd-taskmaster` plugin
8	Depends on: `research-expander` agent (parallel research worker), any research
9	provider configured via `task-master models --set-research` or registered as an
10	MCP research tool.
11
12	## When to Use
13
14	Activate when user says: expand tasks, research tasks, research before coding for all, expand subtasks.
15	Do NOT activate for: single task research (use /research-before-coding), PRD generation (use /prd:go).
16
17	## Native-parallel first (token economy)
18
19	Before launching agent waves, check the cheaper path: the native engine expands tasks
20	in parallel for free. Prefer `python3 script.py expand` — backend op expand (native api) —
21	or the `expand_tasks` MCP tool: it runs structured `expand` across pending tasks
22	concurrently (inheriting the engine's ThreadPoolExecutor) on economy-tier models /
23	keyless host CLIs and merges atomically.
24	Use THIS skill's agent waves when: no provider/CLI is available, native expand reports
25	failures for specific tasks (rerun just those here), or the research must be repo-grounded
26	(agents can read the codebase; native expand cannot).
27
28	## Prerequisites
29
30	- TaskMaster `tasks.json` must exist (run `/prd:go` first)
31	- A research provider is configured — either (a) `task-master models --set-research <model> --<provider>` for any task-master provider family, or (b) an MCP research tool registered in `~/.claude.json` that Claude Code can call directly (for example `mcp__plugin_prd_go__*` tools or an external search/reason MCP)
32	- At least 1 task in `tasks.json`
33
34	---
35
36	## Workflow (5 Steps)
37
38	### Step 1: Preflight
39
40	```bash
41	python3 skills/expand-tasks/script.py read-tasks
42	```
43
44	Returns JSON: `total`, `expanded`, `pending_expansion`, `tasks[]`.
45
46	If `pending_expansion` is 0: Report all tasks already expanded. Exit skill.
47
48	If research provider is not configured: Check via `task-master models` and verify a research role is set. If none, tell the user to configure one (`task-master models --set-research <model> --<provider>`) and exit. The skill does not assume any specific research backend — it uses whatever is configured.
49
50	---
51
52	### Step 2: Choose Scope
53
54	Use AskUserQuestion:
55	- All tasks (default): Expand every task that hasn't been researched yet
56	- Specific tasks: User provides task IDs (comma-separated)
57	- By dependency level: Expand tasks with no dependencies first, then next wave
58
59	AI judgment: Recommend "All tasks" for initial expansion, "By dependency level" for incremental work.
60
61	---
62
63	### Step 3: Generate Research Prompts
64
65	For each task to expand:
66
67	```bash
68	python3 skills/expand-tasks/script.py gen-prompt --task-id <ID>
69	```
70
71	Returns JSON with `prompt` field containing the full research agent prompt.
72
73	AI judgment: Review the auto-generated prompt. Customize research questions if the task needs domain-specific queries. Add project context from the PRD or session-context files if relevant.
74
75	---
76
77	### Step 4: Launch Parallel Research Agents
78
79	Launch research agents in parallel waves. Each wave = up to 5 concurrent agents.
80
81	For each task, spawn a Task agent using the dedicated `research-expander`
82	subagent type (defined in `agents/research-expander.md`):
83
84	```
85	Task(
86	subagent_type: "research-expander",
87	description: "Research Task <ID>: <title>",
88	run_in_background: true,
89	prompt: <prompt from Step 3>
90	)
91	```
92
93	Wave strategy:
94	- Wave 1: Tasks with no dependencies (they inform downstream tasks) — run in parallel
95	- Wave 2: Tasks depending on Wave 1 — run in parallel
96	- Wave 3+: Continue until all tasks covered — run in parallel per wave
97	- Max 5 agents per wave to avoid overwhelming the configured research backend
98
99	Wait for each wave to complete before launching the next. Parallel dispatch
100	only happens WITHIN a wave; waves themselves are serial.
101
102	---
103
104	### Step 5: Collect and Write Results
105
106	As each `research-expander` agent completes, save its research output:
107
108	1. Write agent output to a temp file:
109	```bash
110	cat > /tmp/research-task-<ID>.md <<'EOF'
111	<agent output>
112	EOF
113	```
114
115	2. Write research back to `tasks.json`:
116	```bash
117	python3 skills/expand-tasks/script.py write-research

People who liked this also viewed

6 picks

Type

expand-tasks

582stars57forks

View on GitHub

How to install expand-tasks?

anombyte93/prd-taskmaster/expand-tasks

$npx -y skills add anombyte93/prd-taskmaster --skill expand-tasks

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Expand Tasks with Research v1.0
2
3	Expands TaskMaster tasks with research before coding begins.
4	Deterministic operations handled by `script.py`; AI handles judgment.
5
6	Script location: `skills/expand-tasks/script.py` (relative to plugin root)
7	Part of: `prd-taskmaster` plugin
8	Depends on: `research-expander` agent (parallel research worker), any research
9	provider configured via `task-master models --set-research` or registered as an
10	MCP research tool.
11
12	## When to Use
13
14	Activate when user says: expand tasks, research tasks, research before coding for all, expand subtasks.
15	Do NOT activate for: single task research (use /research-before-coding), PRD generation (use /prd:go).
16
17	## Native-parallel first (token economy)
18
19	Before launching agent waves, check the cheaper path: the native engine expands tasks
20	in parallel for free. Prefer `python3 script.py expand` — backend op expand (native api) —
21	or the `expand_tasks` MCP tool: it runs structured `expand` across pending tasks
22	concurrently (inheriting the engine's ThreadPoolExecutor) on economy-tier models /
23	keyless host CLIs and merges atomically.
24	Use THIS skill's agent waves when: no provider/CLI is available, native expand reports
25	failures for specific tasks (rerun just those here), or the research must be repo-grounded
26	(agents can read the codebase; native expand cannot).
27
28	## Prerequisites
29
30	- TaskMaster `tasks.json` must exist (run `/prd:go` first)
31	- A research provider is configured — either (a) `task-master models --set-research <model> --<provider>` for any task-master provider family, or (b) an MCP research tool registered in `~/.claude.json` that Claude Code can call directly (for example `mcp__plugin_prd_go__*` tools or an external search/reason MCP)
32	- At least 1 task in `tasks.json`
33
34	---
35
36	## Workflow (5 Steps)
37
38	### Step 1: Preflight
39
40	```bash
41	python3 skills/expand-tasks/script.py read-tasks
42	```
43
44	Returns JSON: `total`, `expanded`, `pending_expansion`, `tasks[]`.
45
46	If `pending_expansion` is 0: Report all tasks already expanded. Exit skill.
47
48	If research provider is not configured: Check via `task-master models` and verify a research role is set. If none, tell the user to configure one (`task-master models --set-research <model> --<provider>`) and exit. The skill does not assume any specific research backend — it uses whatever is configured.
49
50	---
51
52	### Step 2: Choose Scope
53
54	Use AskUserQuestion:
55	- All tasks (default): Expand every task that hasn't been researched yet
56	- Specific tasks: User provides task IDs (comma-separated)
57	- By dependency level: Expand tasks with no dependencies first, then next wave
58
59	AI judgment: Recommend "All tasks" for initial expansion, "By dependency level" for incremental work.
60
61	---
62
63	### Step 3: Generate Research Prompts
64
65	For each task to expand:
66
67	```bash
68	python3 skills/expand-tasks/script.py gen-prompt --task-id <ID>
69	```
70
71	Returns JSON with `prompt` field containing the full research agent prompt.
72
73	AI judgment: Review the auto-generated prompt. Customize research questions if the task needs domain-specific queries. Add project context from the PRD or session-context files if relevant.
74
75	---
76
77	### Step 4: Launch Parallel Research Agents
78
79	Launch research agents in parallel waves. Each wave = up to 5 concurrent agents.
80
81	For each task, spawn a Task agent using the dedicated `research-expander`
82	subagent type (defined in `agents/research-expander.md`):
83
84	```
85	Task(
86	subagent_type: "research-expander",
87	description: "Research Task <ID>: <title>",
88	run_in_background: true,
89	prompt: <prompt from Step 3>
90	)
91	```
92
93	Wave strategy:
94	- Wave 1: Tasks with no dependencies (they inform downstream tasks) — run in parallel
95	- Wave 2: Tasks depending on Wave 1 — run in parallel
96	- Wave 3+: Continue until all tasks covered — run in parallel per wave
97	- Max 5 agents per wave to avoid overwhelming the configured research backend
98
99	Wait for each wave to complete before launching the next. Parallel dispatch
100	only happens WITHIN a wave; waves themselves are serial.
101
102	---
103
104	### Step 5: Collect and Write Results
105
106	As each `research-expander` agent completes, save its research output:
107
108	1. Write agent output to a temp file:
109	```bash
110	cat > /tmp/research-task-<ID>.md <<'EOF'
111	<agent output>
112	EOF
113	```
114
115	2. Write research back to `tasks.json`:
116	```bash
117	python3 skills/expand-tasks/script.py write-research

People who liked this also viewed

6 picks

Type