multi-llm-advisor

Name: multi-llm-advisor
Author: Svenja-dev

21installs5forks

How to install multi-llm-advisor?

svenja-dev/claude-code-skills/multi-llm-advisor

$npx -y skills add Svenja-dev/claude-code-skills --skill multi-llm-advisor

Installs into the current project.

Summary

Fetches additional perspectives from OpenAI Codex and Google Gemini for architecture, review, and debugging. Transparently displays all LLM calls.

Files · 1

View on GitHub

SKILL.md

1	# Multi-LLM Advisor
2
3	This skill calls Codex 5.1 Pro and Gemini 3 Pro to provide additional perspectives.
4
5	## When to Activate
6
7	- Architecture decisions (new features, refactoring)
8	- Code review (before commits, PRs)
9	- Debugging (complex errors, performance issues)
10	- On explicit request ("second opinion", "different perspective")
11
12	## Usage
13
14	```
15	Use the multi-llm-advisor skill to get architecture feedback on [topic]
16	Use the multi-llm-advisor skill to review this code
17	Use the multi-llm-advisor skill to debug [issue]
18	```
19
20	## Transparency Format
21
22	Every invocation displays the following:
23
24	```
25	+==============================================================+
26	\| MULTI-LLM ADVISOR - [MODE: ARCHITECTURE\|REVIEW\|DEBUG] \|
27	+==============================================================+
28	\| CONTEXT SENT TO LLMs: \|
29	\| - Files: [list] \|
30	\| - Question: [prompt] \|
31	\| - Tokens: ~[count] \|
32	+--------------------------------------------------------------+
33	\| CODEX 5.1 PRO RESPONSE: \|
34	\| [response] \|
35	+--------------------------------------------------------------+
36	\| GEMINI 3 PRO RESPONSE: \|
37	\| [response] \|
38	+--------------------------------------------------------------+
39	\| SYNTHESIS (Claude's Recommendation): \|
40	\| [combined analysis] \|
41	+==============================================================+
42	```
43
44	## Prompt Templates
45
46	### Architecture Mode
47	```
48	You are a senior software architect. Analyze this architecture decision:
49
50	CONTEXT:
51	{context}
52
53	QUESTION:
54	{question}
55
56	CURRENT STACK:
57	{stack}
58
59	Provide:
60	1. Pros/Cons of the proposed approach
61	2. Alternative approaches (max 2)
62	3. Potential risks and mitigations
63	4. Recommendation with reasoning
64
65	Be concise. Max 300 words.
66	```
67
68	### Review Mode
69	```
70	You are a senior code reviewer. Review this code:
71
72	CODE:
73	{code}
74
75	LANGUAGE: {language}
76	PROJECT TYPE: {project_type}
77
78	Focus on:
79	1. Security vulnerabilities (OWASP Top 10)
80	2. Performance issues
81	3. Maintainability concerns
82	4. TypeScript/type safety (if applicable)
83
84	Format: Bullet points, max 200 words.
85	```
86
87	### Debug Mode
88	```
89	You are a debugging expert. Analyze this issue:
90
91	ERROR/SYMPTOM:
92	{error}
93
94	RELEVANT CODE:
95	{code}
96
97	CONTEXT:
98	{context}
99
100	Provide:
101	1. Root cause analysis (most likely)
102	2. 2-3 diagnostic steps
103	3. Suggested fix with code
104
105	Be specific and actionable. Max 250 words.
106	```
107
108	## API Configuration
109
110	Environment variables (store in `.env` or system env):
111	- `OPENAI_API_KEY` - For Codex 5.1 Pro
112	- `GOOGLE_AI_API_KEY` - For Gemini 3 Pro (same as gemini-image-gen)
113
114	## Script Location
115
116	`~/.claude/skills/multi-llm-advisor/advisor.ts`
117
118	## Hook Integration
119
120	Triggered via `multi-llm-advisor-hook.ts` when:
121	- PreToolUse: Detects architecture/review/debug keywords
122	- Manual: User explicitly requests second opinion
123
124	## Real-World Example
125
126	When building the Gemini API integration at [fabrikIQ.com](https://www.fabrikiq.com), this skill helped decide between:
127	- Vertex AI (enterprise, EU region support) vs Google AI Studio (simpler)
128	- Streaming vs batch responses for large manufacturing datasets
129	- The synthesis recommended Vertex AI for GDPR compliance, which proved correct.

People who liked this also viewed

6 picks

Type

hyperframes-cliUse the HyperFrames CLI development loop: init, add, catalog, capture, lint, check, snapshot, compare, grade-compare, preview, play, present, beats, keyframes, single or batch render, publish, cloud, SkillsJul 2026262k37k
request-refactor-planSkills for Real Engineers. Straight from my .agents directory.SkillsJul 2026165k181k
analyze-projectRigor Analyze / Rigor Audit read-only skill for deep learning research repositories. Use when the user wants to read and understand a repository, inspect model structure and training or inference entrSkillsMay 2026161k507
explore-codeRigor Improve implementation leaf skill for auditable candidate implementation in deep learning research repositories. Use when the researcher explicitly authorizes exploratory work on an isolated braSkillsMay 2026161k507
ubiquitous-languageSkills for Real Engineers. Straight from my .agents directory.SkillsJul 2026160k181k
scaffold-exercisesSkills for Real Engineers. Straight from my .agents directory.SkillsJul 2026149k181k

multi-llm-advisor

21installs5forks

View on GitHub

How to install multi-llm-advisor?

svenja-dev/claude-code-skills/multi-llm-advisor

$npx -y skills add Svenja-dev/claude-code-skills --skill multi-llm-advisor

Installs into the current project.

Summary

Fetches additional perspectives from OpenAI Codex and Google Gemini for architecture, review, and debugging. Transparently displays all LLM calls.

Files · 1

View on GitHub

SKILL.md

1	# Multi-LLM Advisor
2
3	This skill calls Codex 5.1 Pro and Gemini 3 Pro to provide additional perspectives.
4
5	## When to Activate
6
7	- Architecture decisions (new features, refactoring)
8	- Code review (before commits, PRs)
9	- Debugging (complex errors, performance issues)
10	- On explicit request ("second opinion", "different perspective")
11
12	## Usage
13
14	```
15	Use the multi-llm-advisor skill to get architecture feedback on [topic]
16	Use the multi-llm-advisor skill to review this code
17	Use the multi-llm-advisor skill to debug [issue]
18	```
19
20	## Transparency Format
21
22	Every invocation displays the following:
23
24	```
25	+==============================================================+
26	\| MULTI-LLM ADVISOR - [MODE: ARCHITECTURE\|REVIEW\|DEBUG] \|
27	+==============================================================+
28	\| CONTEXT SENT TO LLMs: \|
29	\| - Files: [list] \|
30	\| - Question: [prompt] \|
31	\| - Tokens: ~[count] \|
32	+--------------------------------------------------------------+
33	\| CODEX 5.1 PRO RESPONSE: \|
34	\| [response] \|
35	+--------------------------------------------------------------+
36	\| GEMINI 3 PRO RESPONSE: \|
37	\| [response] \|
38	+--------------------------------------------------------------+
39	\| SYNTHESIS (Claude's Recommendation): \|
40	\| [combined analysis] \|
41	+==============================================================+
42	```
43
44	## Prompt Templates
45
46	### Architecture Mode
47	```
48	You are a senior software architect. Analyze this architecture decision:
49
50	CONTEXT:
51	{context}
52
53	QUESTION:
54	{question}
55
56	CURRENT STACK:
57	{stack}
58
59	Provide:
60	1. Pros/Cons of the proposed approach
61	2. Alternative approaches (max 2)
62	3. Potential risks and mitigations
63	4. Recommendation with reasoning
64
65	Be concise. Max 300 words.
66	```
67
68	### Review Mode
69	```
70	You are a senior code reviewer. Review this code:
71
72	CODE:
73	{code}
74
75	LANGUAGE: {language}
76	PROJECT TYPE: {project_type}
77
78	Focus on:
79	1. Security vulnerabilities (OWASP Top 10)
80	2. Performance issues
81	3. Maintainability concerns
82	4. TypeScript/type safety (if applicable)
83
84	Format: Bullet points, max 200 words.
85	```
86
87	### Debug Mode
88	```
89	You are a debugging expert. Analyze this issue:
90
91	ERROR/SYMPTOM:
92	{error}
93
94	RELEVANT CODE:
95	{code}
96
97	CONTEXT:
98	{context}
99
100	Provide:
101	1. Root cause analysis (most likely)
102	2. 2-3 diagnostic steps
103	3. Suggested fix with code
104
105	Be specific and actionable. Max 250 words.
106	```
107
108	## API Configuration
109
110	Environment variables (store in `.env` or system env):
111	- `OPENAI_API_KEY` - For Codex 5.1 Pro
112	- `GOOGLE_AI_API_KEY` - For Gemini 3 Pro (same as gemini-image-gen)
113
114	## Script Location
115
116	`~/.claude/skills/multi-llm-advisor/advisor.ts`
117
118	## Hook Integration
119
120	Triggered via `multi-llm-advisor-hook.ts` when:
121	- PreToolUse: Detects architecture/review/debug keywords
122	- Manual: User explicitly requests second opinion
123
124	## Real-World Example
125
126	When building the Gemini API integration at [fabrikIQ.com](https://www.fabrikiq.com), this skill helped decide between:
127	- Vertex AI (enterprise, EU region support) vs Google AI Studio (simpler)
128	- Streaming vs batch responses for large manufacturing datasets
129	- The synthesis recommended Vertex AI for GDPR compliance, which proved correct.

People who liked this also viewed

6 picks

Type