tech-spec-planning-coarse

Name: tech-spec-planning-coarse
Author: pavel-molyanov

241stars75forks

How to install tech-spec-planning-coarse?

pavel-molyanov/molyanov-ai-dev/tech-spec-planning-coarse

$npx -y skills add pavel-molyanov/molyanov-ai-dev --skill tech-spec-planning-coarse

Installs into the current project.

Summary

Creates tech-spec.md with architecture, decisions, testing strategy, and coarse implementation plan. Coarse variant: groups related work into skill-scoped tasks instead of atomic ones. Fewer tasks, each covers one skill's cohesive scope in the feature. Use when: "сделай техспек c

Files · 1

View on GitHub

SKILL.md

1	<!-- Generated by sync-to-codex v1. Do not edit directly. -->
2
3	# Tech Spec Planning (Coarse Variant)
4
5	Create technical specification with coarse implementation plan: one task per skill's cohesive scope, not atomic units.
6
7	Input: `work/{feature}/user-spec.md` + Project Knowledge
8	Output: `work/{feature}/tech-spec.md` (approved)
9	Language: Technical documentation in English; communication with the user in the language the user writes in
10
11	## Phase 1: Load Context
12
13	1. Ask user for feature name if not provided. Check `work/{feature}/` exists, create if needed.
14
15	2. Read `work/{feature}/user-spec.md`. If missing — ask user to describe the task or create user-spec first.
16	Extract `size: S\|M\|L` from user-spec frontmatter — it determines testing strategy depth.
17
18	3. Read all files in `.claude/skills/project-knowledge/references/` (project.md, architecture.md, patterns.md, deployment.md, ux-guidelines.md, and any custom domain files). Missing files are fine.
19
20	4. user-spec.md is the single input source — all information from interview.yml and code research is already consolidated there.
21
22	Checkpoint:
23	- [ ] Feature folder exists
24	- [ ] user-spec.md read, size extracted
25	- [ ] Project Knowledge read
26
27	## Phase 2: Code Research
28
29	Launch `code-researcher` subagent (spawn_agent, opus) with feature path and user-spec path. The agent reads existing `code-research.md` (from user-spec phase if available) and deepens analysis for implementation.
30
31	After subagent completes — read `{feature_path}/code-research.md`. Use in Phase 3 clarification and Phase 4 spec writing.
32
33	If during later phases a gap is discovered — launch `code-researcher` again with the specific question.
34
35	Checkpoint:
36	- [ ] code-research.md created/updated with implementation-level analysis
37	- [ ] Research file read by orchestrator
38
39	## Phase 3: Clarification (Adaptive)
40
41	Analyze if additional information is needed based on user-spec and code research.
42
43	- Ask technical questions if gaps exist.
44	- Focus: technical constraints, integration points, data sources, external dependencies.
45	- If gaps found in user-spec requirements — discuss with user and update user-spec too.
46	- If requirements are fundamentally unclear — suggest creating user-spec first.
47
48	Checkpoint:
49	- [ ] All technical gaps clarified (or none existed)
50
51	## Phase 4: Create tech-spec
52
53	1. Copy template to feature folder:
54	```bash
55	cp ~/.claude/shared/work-templates/tech-spec.md.template work/{feature}/tech-spec.md
56	```
57	Then edit sections one by one using Edit tool.
58
59	2. Fill frontmatter:
60	- `created`: today's date
61	- `status`: draft
62	- `size`: copy from user-spec (S\|M\|L)
63	- `branch`: `dev`
64
65	3. Fill all template sections. The template defines section structure — follow it directly.
66	In Architecture → Shared Resources: list heavy resources (ML models, DB pools, browser instances, API clients) shared across components. Specify owner, consumers, instance count. If none — write "None".
67
68	User-spec anchoring: Every decision in the Decisions section must reference a user-spec requirement it serves. If a decision is purely technical — mark it `[TECHNICAL]`. If a decision contradicts user-spec — document it in User-Spec Deviations as `[PENDING USER APPROVAL]`.
69
70	4. Coarse decomposition rule: fill Implementation Tasks by grouping related work into skill-scoped tasks.
71
72	How to decompose:
73	- Start from the list of execution skills needed (from [skills-and-reviewers.md](~/.claude/skills/tech-spec-planning/references/skills-and-reviewers.md)): `code-writing`, `layout-writing`, `prompt-master`, `skill-master`, `infrastructure-setup`, `deploy-pipeline`.
74	- Group into one task: work that shares a skill AND (same module OR shared files OR logical dependency). All ingestion code → one `code-writing` task. All system prompts for one pipeline → one `prompt-master` task.
75	- Split into separate tasks: independent modules within the same skill (enables parallel execution in the same wave). Backend ingest + frontend dashboard → two `code-writing` tasks when each owns a separate set of files and can proceed in parallel.
76	- Layout ownership: pure layout is its own one-skill `layout-writing` task. Planned mixed layout + business-logic scope splits into l

People who liked this also viewed

6 picks

Type

tech-spec-planning-coarse

241stars75forks

View on GitHub

How to install tech-spec-planning-coarse?

pavel-molyanov/molyanov-ai-dev/tech-spec-planning-coarse

$npx -y skills add pavel-molyanov/molyanov-ai-dev --skill tech-spec-planning-coarse

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	<!-- Generated by sync-to-codex v1. Do not edit directly. -->
2
3	# Tech Spec Planning (Coarse Variant)
4
5	Create technical specification with coarse implementation plan: one task per skill's cohesive scope, not atomic units.
6
7	Input: `work/{feature}/user-spec.md` + Project Knowledge
8	Output: `work/{feature}/tech-spec.md` (approved)
9	Language: Technical documentation in English; communication with the user in the language the user writes in
10
11	## Phase 1: Load Context
12
13	1. Ask user for feature name if not provided. Check `work/{feature}/` exists, create if needed.
14
15	2. Read `work/{feature}/user-spec.md`. If missing — ask user to describe the task or create user-spec first.
16	Extract `size: S\|M\|L` from user-spec frontmatter — it determines testing strategy depth.
17
18	3. Read all files in `.claude/skills/project-knowledge/references/` (project.md, architecture.md, patterns.md, deployment.md, ux-guidelines.md, and any custom domain files). Missing files are fine.
19
20	4. user-spec.md is the single input source — all information from interview.yml and code research is already consolidated there.
21
22	Checkpoint:
23	- [ ] Feature folder exists
24	- [ ] user-spec.md read, size extracted
25	- [ ] Project Knowledge read
26
27	## Phase 2: Code Research
28
29	Launch `code-researcher` subagent (spawn_agent, opus) with feature path and user-spec path. The agent reads existing `code-research.md` (from user-spec phase if available) and deepens analysis for implementation.
30
31	After subagent completes — read `{feature_path}/code-research.md`. Use in Phase 3 clarification and Phase 4 spec writing.
32
33	If during later phases a gap is discovered — launch `code-researcher` again with the specific question.
34
35	Checkpoint:
36	- [ ] code-research.md created/updated with implementation-level analysis
37	- [ ] Research file read by orchestrator
38
39	## Phase 3: Clarification (Adaptive)
40
41	Analyze if additional information is needed based on user-spec and code research.
42
43	- Ask technical questions if gaps exist.
44	- Focus: technical constraints, integration points, data sources, external dependencies.
45	- If gaps found in user-spec requirements — discuss with user and update user-spec too.
46	- If requirements are fundamentally unclear — suggest creating user-spec first.
47
48	Checkpoint:
49	- [ ] All technical gaps clarified (or none existed)
50
51	## Phase 4: Create tech-spec
52
53	1. Copy template to feature folder:
54	```bash
55	cp ~/.claude/shared/work-templates/tech-spec.md.template work/{feature}/tech-spec.md
56	```
57	Then edit sections one by one using Edit tool.
58
59	2. Fill frontmatter:
60	- `created`: today's date
61	- `status`: draft
62	- `size`: copy from user-spec (S\|M\|L)
63	- `branch`: `dev`
64
65	3. Fill all template sections. The template defines section structure — follow it directly.
66	In Architecture → Shared Resources: list heavy resources (ML models, DB pools, browser instances, API clients) shared across components. Specify owner, consumers, instance count. If none — write "None".
67
68	User-spec anchoring: Every decision in the Decisions section must reference a user-spec requirement it serves. If a decision is purely technical — mark it `[TECHNICAL]`. If a decision contradicts user-spec — document it in User-Spec Deviations as `[PENDING USER APPROVAL]`.
69
70	4. Coarse decomposition rule: fill Implementation Tasks by grouping related work into skill-scoped tasks.
71
72	How to decompose:
73	- Start from the list of execution skills needed (from [skills-and-reviewers.md](~/.claude/skills/tech-spec-planning/references/skills-and-reviewers.md)): `code-writing`, `layout-writing`, `prompt-master`, `skill-master`, `infrastructure-setup`, `deploy-pipeline`.
74	- Group into one task: work that shares a skill AND (same module OR shared files OR logical dependency). All ingestion code → one `code-writing` task. All system prompts for one pipeline → one `prompt-master` task.
75	- Split into separate tasks: independent modules within the same skill (enables parallel execution in the same wave). Backend ingest + frontend dashboard → two `code-writing` tasks when each owns a separate set of files and can proceed in parallel.
76	- Layout ownership: pure layout is its own one-skill `layout-writing` task. Planned mixed layout + business-logic scope splits into l

People who liked this also viewed

6 picks

Type