code-forge

Name: code-forge
Author: techygarg

165stars12forks

How to install code-forge?

techygarg/lattice/code-forge

$npx -y skills add techygarg/lattice --skill code-forge

Installs into the current project.

Summary

Generate implementation code from an approved design blueprint or verbal requirements. Composes context anchoring, architecture, clean code, DDD, security, and test quality into an inside-out implementation workflow. Use when moving from design to code, implementing approved cont

Files · 1

View on GitHub

SKILL.md

1	# Code Forge
2
3	## Required Skills
4
5	Read, apply:
6
7	1. `framework:knowledge-priming` -- Load proj context (stack, arch, conventions) so impl matches real proj (always)
8	2. `framework:context-anchoring` -- Load/find context anchor doc; enrich as impl decisions made (always)
9	3. `framework:learning-harvest` -- Load prior operational learnings inform impl; harvest new patterns at session end (always)
10	4. `framework:collaborative-judgment` -- Surface real judgment calls w/ structured opts vs silent assume (always)
11	5. `framework:architecture` -- Layer place, dep direction, struct valid (always)
12	6. `framework:clean-code` -- Craft rails: SRP, naming, complexity, err handle (always)
13	7. `framework:domain-driven-design` -- Aggregates, entities, VOs, domain svcs (conditional: only when touch domain folder)
14	8. `framework:secure-coding` -- Trust bounds, injection prevent, secrets mgmt (conditional: only boundary-cross code)
15	9. `framework:test-quality` -- AAA struct, isolation, assert quality, naming (always when write tests)
16
17	## Workflow
18
19	### Step 1: Establish Implementation Context
20
21	Use `framework:learning-harvest` Load behavior. Focus hint: "implementation session — focus: implementation craft, quality signals, reliability".
22
23	Use `framework:context-anchoring` Doc Discovery check existing context anchor doc for feature impl.
24
25	- If found → Load (context-anchor Load behavior). Present struct ack -- feature name, status, decision count, open Qs, constraints. Honor all logged decisions/constraints as active commits.
26	- If not found → Nudge user: "Have design doc/blueprint for feature? Or work from discussed?" Accept either graceful.
27	- User provides doc → load, follow.
28	- Proceed without → all atom rails still apply; just no approved design doc to reference. Work from verbal reqs in convo.
29
30	Design completeness check — STOP gates before Step 2:
31
32	Check 1 — status: Read frontmatter `status`.
33	- `approved` → pass.
34	- Anything else → STOP: "Context doc not approved (`status: [value]`). Run design-blueprint first. Proceed anyway?" Confirm → log in Decisions Log, continue as "Without approved design."
35
36	Check 2 — levels present: Scan body for `## Design: Level 3` and `## Design: Level 4`.
37	- Both present → pass.
38	- Either missing → STOP: "Missing [Level 3 / Level 4 / both]. Proceed anyway?" Confirm → log absent levels in Decisions Log, treat as gaps to fill during implementation.
39
40	Both pass → proceed as "With approved design".
41
42	### Step 2: Plan Implementation Order
43
44	With approved design: Extract component list, layer assigns from context anchor doc. Use L2 (Components) decisions for layer place, L3 (Interactions) for dep flow.
45
46	Without approved design: Classify req components→arch layers using layer defs from `framework:architecture`. Each component, determine:
47
48	- Primary responsibility? (biz rules, data access, coord, external I/O)
49	- Which layer in loaded arch doc matches responsibility?
50	- Dep constraints for that layer?
51
52	If `framework:architecture` no loaded layer defs (neither defaults nor custom doc resolved), warn: "No arch rules avail. Run `/architecture-refiner` define arch standards. Proceed w/o arch guidance." Continue w/ only remaining atom rails.
53
54	Present proposed layer assigns→user for approval before proceed.
55
56	Both cases, plan inside-out impl order following dep direction from loaded arch doc — start innermost layer (no outward deps), work outward. Each layer's deps should exist when built.
57
58	Classify each op per flow patterns in loaded arch doc (e.g., cmd vs query flows, or equiv distinction your arch style).
59
60	Present impl plan -- ordered component list, layer assigns, flow classifs -- confirm w/ user before write code.
61
62	After plan approved, ask user choose review mode:
63
64	> "How review impl?"
65	> 1. Layer-by-layer (rec) -- Impl each layer fully, pause for review before next. One review pt/layer.
66	> 2. Full autonomy -- Impl everything end-to-end, present complete result. One review pt at end. (If blueprint exists, still pause any deviation from approved design.)
67	> 3. Component-by-component -- Pause after each individual component for feedback. Max review pts.
68
69	Default layer-by-layer if user no preference.
70
71	### Step 3: Implement Per Component
72
73	Each component in planned order, gen code+tests together -- tests not afterthought.
74
75	Every component:
76
77	- Prefer the simpler path first. Before writing custom code: does a stdlib function, platform built-in, or existing dependency already cover this? If yes — use it. Can it be expressed in shorter code? Use

People who liked this also viewed

6 picks

Type

code-forge

165stars12forks

View on GitHub

How to install code-forge?

techygarg/lattice/code-forge

$npx -y skills add techygarg/lattice --skill code-forge

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Code Forge
2
3	## Required Skills
4
5	Read, apply:
6
7	1. `framework:knowledge-priming` -- Load proj context (stack, arch, conventions) so impl matches real proj (always)
8	2. `framework:context-anchoring` -- Load/find context anchor doc; enrich as impl decisions made (always)
9	3. `framework:learning-harvest` -- Load prior operational learnings inform impl; harvest new patterns at session end (always)
10	4. `framework:collaborative-judgment` -- Surface real judgment calls w/ structured opts vs silent assume (always)
11	5. `framework:architecture` -- Layer place, dep direction, struct valid (always)
12	6. `framework:clean-code` -- Craft rails: SRP, naming, complexity, err handle (always)
13	7. `framework:domain-driven-design` -- Aggregates, entities, VOs, domain svcs (conditional: only when touch domain folder)
14	8. `framework:secure-coding` -- Trust bounds, injection prevent, secrets mgmt (conditional: only boundary-cross code)
15	9. `framework:test-quality` -- AAA struct, isolation, assert quality, naming (always when write tests)
16
17	## Workflow
18
19	### Step 1: Establish Implementation Context
20
21	Use `framework:learning-harvest` Load behavior. Focus hint: "implementation session — focus: implementation craft, quality signals, reliability".
22
23	Use `framework:context-anchoring` Doc Discovery check existing context anchor doc for feature impl.
24
25	- If found → Load (context-anchor Load behavior). Present struct ack -- feature name, status, decision count, open Qs, constraints. Honor all logged decisions/constraints as active commits.
26	- If not found → Nudge user: "Have design doc/blueprint for feature? Or work from discussed?" Accept either graceful.
27	- User provides doc → load, follow.
28	- Proceed without → all atom rails still apply; just no approved design doc to reference. Work from verbal reqs in convo.
29
30	Design completeness check — STOP gates before Step 2:
31
32	Check 1 — status: Read frontmatter `status`.
33	- `approved` → pass.
34	- Anything else → STOP: "Context doc not approved (`status: [value]`). Run design-blueprint first. Proceed anyway?" Confirm → log in Decisions Log, continue as "Without approved design."
35
36	Check 2 — levels present: Scan body for `## Design: Level 3` and `## Design: Level 4`.
37	- Both present → pass.
38	- Either missing → STOP: "Missing [Level 3 / Level 4 / both]. Proceed anyway?" Confirm → log absent levels in Decisions Log, treat as gaps to fill during implementation.
39
40	Both pass → proceed as "With approved design".
41
42	### Step 2: Plan Implementation Order
43
44	With approved design: Extract component list, layer assigns from context anchor doc. Use L2 (Components) decisions for layer place, L3 (Interactions) for dep flow.
45
46	Without approved design: Classify req components→arch layers using layer defs from `framework:architecture`. Each component, determine:
47
48	- Primary responsibility? (biz rules, data access, coord, external I/O)
49	- Which layer in loaded arch doc matches responsibility?
50	- Dep constraints for that layer?
51
52	If `framework:architecture` no loaded layer defs (neither defaults nor custom doc resolved), warn: "No arch rules avail. Run `/architecture-refiner` define arch standards. Proceed w/o arch guidance." Continue w/ only remaining atom rails.
53
54	Present proposed layer assigns→user for approval before proceed.
55
56	Both cases, plan inside-out impl order following dep direction from loaded arch doc — start innermost layer (no outward deps), work outward. Each layer's deps should exist when built.
57
58	Classify each op per flow patterns in loaded arch doc (e.g., cmd vs query flows, or equiv distinction your arch style).
59
60	Present impl plan -- ordered component list, layer assigns, flow classifs -- confirm w/ user before write code.
61
62	After plan approved, ask user choose review mode:
63
64	> "How review impl?"
65	> 1. Layer-by-layer (rec) -- Impl each layer fully, pause for review before next. One review pt/layer.
66	> 2. Full autonomy -- Impl everything end-to-end, present complete result. One review pt at end. (If blueprint exists, still pause any deviation from approved design.)
67	> 3. Component-by-component -- Pause after each individual component for feedback. Max review pts.
68
69	Default layer-by-layer if user no preference.
70
71	### Step 3: Implement Per Component
72
73	Each component in planned order, gen code+tests together -- tests not afterthought.
74
75	Every component:
76
77	- Prefer the simpler path first. Before writing custom code: does a stdlib function, platform built-in, or existing dependency already cover this? If yes — use it. Can it be expressed in shorter code? Use

People who liked this also viewed

6 picks

Type