build-advisor

Name: build-advisor
Author: Undertone0809

247stars30forks

How to install build-advisor?

undertone0809/rudder/build-advisor

$npx -y skills add Undertone0809/rudder --skill build-advisor

Installs into the current project.

Summary

Expert advisor for when a build, UI, workflow, spec, or implementation feels wrong but the user cannot yet express the right product, design, engineering, or evaluation critique. Use before more implementation to turn vague dissatisfaction, weak AI-built results, traces, benchmar

Files · 1

View on GitHub

SKILL.md

1	# Build Advisor
2
3	This skill exists for the moment after "something was built" but before the user has a clean professional critique.
4
5	It is not an implementation skill first.
6	It is a diagnosis, proposal, and routing skill.
7
8	Use it when the user needs an expert advisor to turn fuzzy discomfort into:
9
10	- a clearer problem statement
11	- a professional diagnosis
12	- a first-principles map of user scenarios, needs, non-needs, and corner cases
13	- explicit evaluation criteria
14	- 2-3 realistic options
15	- one decision-ready proposal for the recommended option
16	- one recommended next move
17
18	## What This Skill Does
19
20	This skill acts like a cross-functional advisor spanning:
21
22	- product framing
23	- UX and visual design
24	- information architecture
25	- engineering shape
26	- observability and evaluation quality
27	- workflow and process quality
28
29	Its main job is to identify which layer is actually broken.
30
31	Examples:
32
33	- "This UI feels wrong, but I don't know how to explain why."
34	- "You built a version, but it's not good. Help me critique it professionally."
35	- "Before you keep coding, research best practices and tell me what we're missing."
36	- "Should we keep patching this or write a design or architecture doc first?"
37	- "I know the result is too big / too noisy / too complicated, but I need a concrete proposal."
38	- "The trace or benchmark says one thing, but the result still feels wrong. Help me make sense of it."
39
40	## What This Skill Does Not Do
41
42	Do not treat this as a direct code-writing skill by default.
43
44	It should not:
45
46	- jump into implementation before diagnosis
47	- claim "I understand" from the prompt alone when local evidence can be checked
48	- pretend every problem is a UI styling issue
49	- replace specialized execution skills when the right next step is obvious
50	- produce vague "looks better / feels cleaner" advice without criteria
51	- start from the visible implementation detail when the user is asking about
52	the underlying scenario, job-to-be-done, or workflow pressure
53
54	If the correct outcome is to invoke or recommend a more specialized skill, say so clearly.
55
56	## Distinguish From Nearby Skills
57
58	- `office-hours`: use for idea-stage or pre-build product thinking
59	- `rudder-gstack-guide`: use for choosing a gstack chain in the Rudder repo
60	- `design-guide`: use for Rudder UI conventions once the problem is already known
61	- `design-review`: use when the main need is visual QA and polish on a live surface
62	- `plan-eng-review`: use when architecture and execution planning are the main concern
63	- `investigate`: use when the issue is primarily a bug or regression with unclear root cause
64
65	Use `build-advisor` when the user is blocked on judgment, articulation, or deciding which layer of the problem to fix first.
66
67	## Default Workflow
68
69	Follow this sequence unless the user explicitly narrows the task.
70
71	### 0. Mode Gate
72
73	Before doing a full advisor pass, classify the user's immediate intent:
74
75	- `quick_take`: the user asks for a fast sanity check or says not to write a
76	long plan. Keep the answer short and name the main risk plus next move.
77	- `understanding_check`: the user asks "你懂吗" / "先说说". Do a small evidence
78	pass, then reframe the need. Do not implement or write a full proposal yet.
79	- `proposal`: the user asks for a proposal, options, plan, or first-principles
80	analysis. Stay in advisor mode and produce the requested decision artifact.
81	- `visual_options`: the user is reacting to screenshots or asks for UI schemes.
82	Produce concrete visual alternatives such as ASCII wireframes, low-fidelity
83	HTML, or screenshot-backed inspection criteria before recommending code.
84	- `implementation_handoff`: the user approves an option with phrases like
85	"可以，开始推进", "按方案 A 改", or "follow 你的思路，优化一下". Switch out of
86	advisor-only mode and execute normally with repository validation rules.
87	- `reviewer_loop`: the user asks for reviewer rounds, acceptance gates, or
88	independent review. Route to the reviewer-loop workflow instead of acting as
89	a single advisor.
90
91	State the mode only when it clarifies the response. The purpose of the gate is
92	to prevent two failures: writing a long proposal when the user wants a quick
93	decision, and starting implementation when the user asked only for judgment.
94
95	### Plan Template Reference
96
97	When this skill writes or prepares a plan document, read
98	`references/plan-doc-templates.md` before drafting the file. That reference
99	maps proposal and implementation work to the canonical repo templates.
100
101	### 1. Evidence Intake Before Reframing
102
103	Do a small, targeted context

People who liked this also viewed

6 picks

Type

build-advisor

247stars30forks

View on GitHub

How to install build-advisor?

undertone0809/rudder/build-advisor

$npx -y skills add Undertone0809/rudder --skill build-advisor

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Build Advisor
2
3	This skill exists for the moment after "something was built" but before the user has a clean professional critique.
4
5	It is not an implementation skill first.
6	It is a diagnosis, proposal, and routing skill.
7
8	Use it when the user needs an expert advisor to turn fuzzy discomfort into:
9
10	- a clearer problem statement
11	- a professional diagnosis
12	- a first-principles map of user scenarios, needs, non-needs, and corner cases
13	- explicit evaluation criteria
14	- 2-3 realistic options
15	- one decision-ready proposal for the recommended option
16	- one recommended next move
17
18	## What This Skill Does
19
20	This skill acts like a cross-functional advisor spanning:
21
22	- product framing
23	- UX and visual design
24	- information architecture
25	- engineering shape
26	- observability and evaluation quality
27	- workflow and process quality
28
29	Its main job is to identify which layer is actually broken.
30
31	Examples:
32
33	- "This UI feels wrong, but I don't know how to explain why."
34	- "You built a version, but it's not good. Help me critique it professionally."
35	- "Before you keep coding, research best practices and tell me what we're missing."
36	- "Should we keep patching this or write a design or architecture doc first?"
37	- "I know the result is too big / too noisy / too complicated, but I need a concrete proposal."
38	- "The trace or benchmark says one thing, but the result still feels wrong. Help me make sense of it."
39
40	## What This Skill Does Not Do
41
42	Do not treat this as a direct code-writing skill by default.
43
44	It should not:
45
46	- jump into implementation before diagnosis
47	- claim "I understand" from the prompt alone when local evidence can be checked
48	- pretend every problem is a UI styling issue
49	- replace specialized execution skills when the right next step is obvious
50	- produce vague "looks better / feels cleaner" advice without criteria
51	- start from the visible implementation detail when the user is asking about
52	the underlying scenario, job-to-be-done, or workflow pressure
53
54	If the correct outcome is to invoke or recommend a more specialized skill, say so clearly.
55
56	## Distinguish From Nearby Skills
57
58	- `office-hours`: use for idea-stage or pre-build product thinking
59	- `rudder-gstack-guide`: use for choosing a gstack chain in the Rudder repo
60	- `design-guide`: use for Rudder UI conventions once the problem is already known
61	- `design-review`: use when the main need is visual QA and polish on a live surface
62	- `plan-eng-review`: use when architecture and execution planning are the main concern
63	- `investigate`: use when the issue is primarily a bug or regression with unclear root cause
64
65	Use `build-advisor` when the user is blocked on judgment, articulation, or deciding which layer of the problem to fix first.
66
67	## Default Workflow
68
69	Follow this sequence unless the user explicitly narrows the task.
70
71	### 0. Mode Gate
72
73	Before doing a full advisor pass, classify the user's immediate intent:
74
75	- `quick_take`: the user asks for a fast sanity check or says not to write a
76	long plan. Keep the answer short and name the main risk plus next move.
77	- `understanding_check`: the user asks "你懂吗" / "先说说". Do a small evidence
78	pass, then reframe the need. Do not implement or write a full proposal yet.
79	- `proposal`: the user asks for a proposal, options, plan, or first-principles
80	analysis. Stay in advisor mode and produce the requested decision artifact.
81	- `visual_options`: the user is reacting to screenshots or asks for UI schemes.
82	Produce concrete visual alternatives such as ASCII wireframes, low-fidelity
83	HTML, or screenshot-backed inspection criteria before recommending code.
84	- `implementation_handoff`: the user approves an option with phrases like
85	"可以，开始推进", "按方案 A 改", or "follow 你的思路，优化一下". Switch out of
86	advisor-only mode and execute normally with repository validation rules.
87	- `reviewer_loop`: the user asks for reviewer rounds, acceptance gates, or
88	independent review. Route to the reviewer-loop workflow instead of acting as
89	a single advisor.
90
91	State the mode only when it clarifies the response. The purpose of the gate is
92	to prevent two failures: writing a long proposal when the user wants a quick
93	decision, and starting implementation when the user asked only for judgment.
94
95	### Plan Template Reference
96
97	When this skill writes or prepares a plan document, read
98	`references/plan-doc-templates.md` before drafting the file. That reference
99	maps proposal and implementation work to the canonical repo templates.
100
101	### 1. Evidence Intake Before Reframing
102
103	Do a small, targeted context

People who liked this also viewed

6 picks

Type