customise-workflow

Name: customise-workflow
Author: anombyte93

582stars57forks

How to install customise-workflow?

anombyte93/prd-taskmaster/customise-workflow

$npx -y skills add anombyte93/prd-taskmaster --skill customise-workflow

Installs into the current project.

Summary

Customise the prd-taskmaster plugin workflow via curated brainstorm questions. The AI asks, the user answers in plain English, and the skill writes their preferences to .atlas-ai/config/atlas.json. Future runs of prd-taskmaster read that file and apply user preferences to phase g

Files · 1

View on GitHub

SKILL.md

1	# customise-workflow
2
3	AI-driven workflow customisation for the `prd-taskmaster` plugin.
4	Replaces manual JSON editing. Part of the plugin's companion-skills family.
5
6	Script: `skills/customise-workflow/script.py` (all commands output JSON)
7	Plugin config root: `.atlas-ai/` (per-project, lives alongside TaskMaster's
8	`.taskmaster/`)
9
10	## When to Use
11
12	Activate when the user says: "customise workflow", "customize workflow",
13	"adjust PRD settings", "tune the skill", "change my defaults", or
14	"personalise prd-taskmaster".
15
16	Skip: generating a new PRD (use `/prd:go`), executing tasks (use
17	HANDOFF modes), or running research expansion (use `/expand-tasks`).
18
19	## The One Rule
20
21	**The AI asks the questions and writes the config. The user never manually
22	edits JSON. The config file is the output, not the input.** If the user wants
23	tweaks beyond the curated questions, point them at `.atlas-ai/customizations/`
24	(see "Customizations directory" below) — do not hand them raw JSON.
25
26	## Flow
27
28	```
29	LOAD → ASK → VALIDATE → WRITE → VERIFY
30	```
31
32	### Phase 1: LOAD current config
33
34	Run the script to load existing preferences (or defaults if first run):
35
36	```bash
37	python3 skills/customise-workflow/script.py load-config
38	```
39
40	Returns JSON with current preferences across 6 categories: provider,
41	validation, execution, template, autonomous, gates. Writes to
42	`.atlas-ai/config/atlas.json` if missing, seeding defaults.
43
44	### Phase 2: ASK curated questions
45
46	Read `questions/curated-questions.md` and ask each one via `AskUserQuestion`.
47	The questions are curated so plain-English answers map cleanly to config keys.
48	Example:
49
50	```
51	Q1: Which AI provider do you prefer for task generation?
52	Options: Gemini (free, token-efficient), Claude Code (free, Max only),
53	OpenAI GPT-4, Anthropic Direct API, OpenRouter, Ollama (local)
54
55	Q2: How strict should PRD validation be?
56	Options: Strict (block on NEEDS_WORK), Normal (warn but allow GOOD+),
57	Lenient (accept ACCEPTABLE+)
58
59	Q3: Which execution mode should prd-taskmaster default to?
60	Options: A (Plan Mode), B (Ralph loop), C (Atlas Fleet), ...
61
62	...
63	```
64
65	Do NOT ask all questions at once. Ask one curated question at a time and
66	adapt follow-ups based on answers. (Same pattern as
67	`superpowers:brainstorming`.)
68
69	### Phase 3: VALIDATE answers
70
71	Run the script with each user answer as it arrives. The script validates the
72	answer against allowed values and returns either `ok: true` or a hint about
73	what's wrong.
74
75	```bash
76	python3 skills/customise-workflow/script.py validate-answer \
77	--key provider_main --value gemini-cli
78	```
79
80	If validation fails, re-ask the question with the hint. Never write an invalid
81	value.
82
83	### Phase 4: WRITE config
84
85	After all curated questions are answered, commit the config:
86
87	```bash
88	python3 skills/customise-workflow/script.py write-config --input /tmp/answers.json
89	```
90
91	This writes to `.atlas-ai/config/atlas.json` in the current project.
92	Idempotent — re-running customise-workflow reads and updates the existing
93	file. The script creates the `.atlas-ai/config/` directory if missing.
94
95	### Phase 5: VERIFY
96
97	Show the user their final config and confirm it matches their intent:
98
99	```bash
100	python3 skills/customise-workflow/script.py show-config
101	```
102
103	If the user says "that's not what I meant" for any key, re-enter Phase 2 for
104	just that key, re-validate, and re-write.
105
106	## Script Commands Reference
107
108	\| Command \| Purpose \|
109	\|---\|---\|
110	\| `load-config` \| Load current `.atlas-ai/config/atlas.json` (or defaults) \|
111	\| `list-questions` \| Return the curated question set as JSON \|
112	\| `validate-answer --key K --value V` \| Validate a single answer \|
113	\| `write-config --input <file>` \| Write validated answers to `.atlas-ai/config/atlas.json` \|
114	\| `show-config` \| Display current config \|
115	\| `reset-config` \| Delete `.atlas-ai/config/atlas.json` (back to defaults) \|
116
117	## Config Schema
118
119	`.atlas-ai/config/atlas.json` has 7 top-level keys:
120
121	```json
122	{
123	"token_economy": "conservative\|balanced\|performance",
124	"provider": {
125	"main": "gemini-cli\|claude-code\|anthropic\|openai\|openrouter\|ollama\|...",
126	"model_main": "gemini-3-

People who liked this also viewed

6 picks

Type

customise-workflow

582stars57forks

View on GitHub

How to install customise-workflow?

anombyte93/prd-taskmaster/customise-workflow

$npx -y skills add anombyte93/prd-taskmaster --skill customise-workflow

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# customise-workflow
2
3	AI-driven workflow customisation for the `prd-taskmaster` plugin.
4	Replaces manual JSON editing. Part of the plugin's companion-skills family.
5
6	Script: `skills/customise-workflow/script.py` (all commands output JSON)
7	Plugin config root: `.atlas-ai/` (per-project, lives alongside TaskMaster's
8	`.taskmaster/`)
9
10	## When to Use
11
12	Activate when the user says: "customise workflow", "customize workflow",
13	"adjust PRD settings", "tune the skill", "change my defaults", or
14	"personalise prd-taskmaster".
15
16	Skip: generating a new PRD (use `/prd:go`), executing tasks (use
17	HANDOFF modes), or running research expansion (use `/expand-tasks`).
18
19	## The One Rule
20
21	**The AI asks the questions and writes the config. The user never manually
22	edits JSON. The config file is the output, not the input.** If the user wants
23	tweaks beyond the curated questions, point them at `.atlas-ai/customizations/`
24	(see "Customizations directory" below) — do not hand them raw JSON.
25
26	## Flow
27
28	```
29	LOAD → ASK → VALIDATE → WRITE → VERIFY
30	```
31
32	### Phase 1: LOAD current config
33
34	Run the script to load existing preferences (or defaults if first run):
35
36	```bash
37	python3 skills/customise-workflow/script.py load-config
38	```
39
40	Returns JSON with current preferences across 6 categories: provider,
41	validation, execution, template, autonomous, gates. Writes to
42	`.atlas-ai/config/atlas.json` if missing, seeding defaults.
43
44	### Phase 2: ASK curated questions
45
46	Read `questions/curated-questions.md` and ask each one via `AskUserQuestion`.
47	The questions are curated so plain-English answers map cleanly to config keys.
48	Example:
49
50	```
51	Q1: Which AI provider do you prefer for task generation?
52	Options: Gemini (free, token-efficient), Claude Code (free, Max only),
53	OpenAI GPT-4, Anthropic Direct API, OpenRouter, Ollama (local)
54
55	Q2: How strict should PRD validation be?
56	Options: Strict (block on NEEDS_WORK), Normal (warn but allow GOOD+),
57	Lenient (accept ACCEPTABLE+)
58
59	Q3: Which execution mode should prd-taskmaster default to?
60	Options: A (Plan Mode), B (Ralph loop), C (Atlas Fleet), ...
61
62	...
63	```
64
65	Do NOT ask all questions at once. Ask one curated question at a time and
66	adapt follow-ups based on answers. (Same pattern as
67	`superpowers:brainstorming`.)
68
69	### Phase 3: VALIDATE answers
70
71	Run the script with each user answer as it arrives. The script validates the
72	answer against allowed values and returns either `ok: true` or a hint about
73	what's wrong.
74
75	```bash
76	python3 skills/customise-workflow/script.py validate-answer \
77	--key provider_main --value gemini-cli
78	```
79
80	If validation fails, re-ask the question with the hint. Never write an invalid
81	value.
82
83	### Phase 4: WRITE config
84
85	After all curated questions are answered, commit the config:
86
87	```bash
88	python3 skills/customise-workflow/script.py write-config --input /tmp/answers.json
89	```
90
91	This writes to `.atlas-ai/config/atlas.json` in the current project.
92	Idempotent — re-running customise-workflow reads and updates the existing
93	file. The script creates the `.atlas-ai/config/` directory if missing.
94
95	### Phase 5: VERIFY
96
97	Show the user their final config and confirm it matches their intent:
98
99	```bash
100	python3 skills/customise-workflow/script.py show-config
101	```
102
103	If the user says "that's not what I meant" for any key, re-enter Phase 2 for
104	just that key, re-validate, and re-write.
105
106	## Script Commands Reference
107
108	\| Command \| Purpose \|
109	\|---\|---\|
110	\| `load-config` \| Load current `.atlas-ai/config/atlas.json` (or defaults) \|
111	\| `list-questions` \| Return the curated question set as JSON \|
112	\| `validate-answer --key K --value V` \| Validate a single answer \|
113	\| `write-config --input <file>` \| Write validated answers to `.atlas-ai/config/atlas.json` \|
114	\| `show-config` \| Display current config \|
115	\| `reset-config` \| Delete `.atlas-ai/config/atlas.json` (back to defaults) \|
116
117	## Config Schema
118
119	`.atlas-ai/config/atlas.json` has 7 top-level keys:
120
121	```json
122	{
123	"token_economy": "conservative\|balanced\|performance",
124	"provider": {
125	"main": "gemini-cli\|claude-code\|anthropic\|openai\|openrouter\|ollama\|...",
126	"model_main": "gemini-3-

People who liked this also viewed

6 picks

Type