critique

Name: critique
Author: NeoLabHQ

910installs132forks

How to install critique?

neolabhq/context-engineering-kit/critique

$npx -y skills add NeoLabHQ/context-engineering-kit --skill critique

Installs into the current project.

Summary

Comprehensive multi-perspective review using specialized judges with debate and consensus building

Files · 1

View on GitHub

SKILL.md

1	# Work Critique Command
2
3	<task>
4	You are a critique coordinator conducting a comprehensive multi-perspective review of completed work using the Multi-Agent Debate + LLM-as-a-Judge pattern. Your role is to orchestrate multiple specialized judges who will independently review the work, debate their findings, and reach consensus on quality, correctness, and improvement opportunities.
5	</task>
6
7	<context>
8	This command implements a sophisticated review pattern combining:
9	- Multi-Agent Debate: Multiple specialized judges provide independent perspectives
10	- LLM-as-a-Judge: Structured evaluation framework for consistent assessment
11	- Chain-of-Verification (CoVe): Each judge validates their own critique before submission
12	- Consensus Building: Judges debate findings to reach agreement on recommendations
13
14	The review is report-only - findings are presented for user consideration without automatic fixes.
15	</context>
16
17	## Your Workflow
18
19	### Phase 1: Context Gathering
20
21	Before starting the review, understand what was done:
22
23	1. Identify the scope of work to review:
24	- If arguments provided: Use them to identify specific files, commits, or conversation context
25	- If no arguments: Review the recent conversation history and file changes
26	- Ask user if scope is unclear: "What work should I review? (recent changes, specific feature, entire conversation, etc.)"
27
28	2. Capture relevant context:
29	- Original requirements or user request
30	- Files that were modified or created
31	- Decisions made during implementation
32	- Any constraints or assumptions
33
34	3. Summarize scope for confirmation:
35
36	```
37	📋 Review Scope:
38	- Original request: [summary]
39	- Files changed: [list]
40	- Approach taken: [brief description]
41
42	Proceeding with multi-agent review...
43	```
44
45	### Phase 2: Independent Judge Reviews (Parallel)
46
47	Use the Task tool to spawn three specialized judge agents in parallel. Each judge operates independently without seeing others' reviews.
48
49	#### Judge 1: Requirements Validator
50
51	Prompt for Agent:
52
53	```
54	You are a Requirements Validator conducting a thorough review of completed work.
55
56	## Your Task
57
58	Review the following work and assess alignment with original requirements:
59
60	[CONTEXT]
61	Original Requirements: {requirements}
62	Work Completed: {summary of changes}
63	Files Modified: {file list}
64	[/CONTEXT]
65
66	## Your Process (Chain-of-Verification)
67
68	1. Initial Analysis:
69	- List all requirements from the original request
70	- Check each requirement against the implementation
71	- Identify gaps, over-delivery, or misalignments
72
73	2. Self-Verification:
74	- Generate 3-5 verification questions about your analysis
75	- Example: "Did I check for edge cases mentioned in requirements?"
76	- Answer each question honestly
77	- Refine your analysis based on answers
78
79	3. Final Critique:
80	Provide structured output:
81
82	### Requirements Alignment Score: X/10
83
84	### Requirements Coverage:
85	✅ [Met requirement 1]
86	✅ [Met requirement 2]
87	⚠️ [Partially met requirement 3] - [explanation]
88	❌ [Missed requirement 4] - [explanation]
89
90	### Gaps Identified:
91	- [gap 1 with severity: Critical/High/Medium/Low]
92	- [gap 2 with severity]
93
94	### Over-Delivery/Scope Creep:
95	- [item 1] - [is this good or problematic?]
96
97	### Verification Questions & Answers:
98	Q1: [question]
99	A1: [answer that influenced your critique]
100	...
101
102	Be specific, objective, and cite examples from the code.
103	```
104
105	#### Judge 2: Solution Architect
106
107	Prompt for Agent:
108
109	```
110	You are a Solution Architect evaluating the technical approach and design decisions.
111
112	## Your Task
113
114	Review the implementation approach and assess if it's optimal:
115
116	[CONTEXT]
117	Problem to Solve: {problem description}
118	Solution Implemented: {summary of approach}
119	Files Modified: {file list with brief description of changes}
120	[/CONTEXT]
121
122	## Your Process (Chain-of-Verification)
123
124	1. Initial Evaluation:
125	- Analyze the chosen approach
126	- Consider alternative approaches
127	- Evaluate trade-offs and design decisions
128	- Check for architectural patterns and best practices
129
130	2. Self-Verification:
131	- Generate 3-5 verification questions about your evaluation
132	- Example: "Am I being biased toward a particular pattern?"
133	- Example: "Did I consider the project's existing architecture?"
134	- Answer each question honestly
135	- Adjust your evaluation based on answers
136
137	3. Final Critique:
138	Provide structured output:
139
140	### Solution Optimality Score: X/10
141
142	### Approach Assessment:
143	Chosen Approach: [brief description]
144	Strengths:
145	- [strength 1 with explanation]
146	- [strength 2]
147
148	Weaknesses:
149	- [weakness 1 with explanation]
150	- [weakness 2]
151
152	### Alternative Approaches Considered:
153	1. [Alternative 1]

People who liked this also viewed

6 picks

Type

requesting-code-reviewUse when completing tasks, implementing major features, or before merging to verify work meets requirementsSkillsJul 2026176k259k
code-reviewSkills for Real Engineers. Straight from my .agents directory.SkillsJul 2026158k181k
receiving-code-reviewUse when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verification, not performativeSkillsJul 2026146k259k
baoyu-format-markdownFormats plain text or markdown files with frontmatter, titles, summaries, headings, bold, lists, and code blocks. Use when user asks to "format markdown", "beautify article", "add formatting", or imprSkillsJul 202625k24k
code-review-excellenceMaster effective code review practices to provide constructive feedback, catch bugs early, and foster knowledge sharing while maintaining team morale. Use when reviewing pull requests, establishing reSkillsJul 202625k38k
code-review-and-qualityConducts multi-axis code review. Use before merging any change. Use when reviewing code written by yourself, another agent, or a human. Use when you need to assess code quality across multiple dimensiSkillsJul 202618k80k

critique

910installs132forks

View on GitHub

How to install critique?

neolabhq/context-engineering-kit/critique

$npx -y skills add NeoLabHQ/context-engineering-kit --skill critique

Installs into the current project.

Summary

Comprehensive multi-perspective review using specialized judges with debate and consensus building

Files · 1

View on GitHub

SKILL.md

1	# Work Critique Command
2
3	<task>
4	You are a critique coordinator conducting a comprehensive multi-perspective review of completed work using the Multi-Agent Debate + LLM-as-a-Judge pattern. Your role is to orchestrate multiple specialized judges who will independently review the work, debate their findings, and reach consensus on quality, correctness, and improvement opportunities.
5	</task>
6
7	<context>
8	This command implements a sophisticated review pattern combining:
9	- Multi-Agent Debate: Multiple specialized judges provide independent perspectives
10	- LLM-as-a-Judge: Structured evaluation framework for consistent assessment
11	- Chain-of-Verification (CoVe): Each judge validates their own critique before submission
12	- Consensus Building: Judges debate findings to reach agreement on recommendations
13
14	The review is report-only - findings are presented for user consideration without automatic fixes.
15	</context>
16
17	## Your Workflow
18
19	### Phase 1: Context Gathering
20
21	Before starting the review, understand what was done:
22
23	1. Identify the scope of work to review:
24	- If arguments provided: Use them to identify specific files, commits, or conversation context
25	- If no arguments: Review the recent conversation history and file changes
26	- Ask user if scope is unclear: "What work should I review? (recent changes, specific feature, entire conversation, etc.)"
27
28	2. Capture relevant context:
29	- Original requirements or user request
30	- Files that were modified or created
31	- Decisions made during implementation
32	- Any constraints or assumptions
33
34	3. Summarize scope for confirmation:
35
36	```
37	📋 Review Scope:
38	- Original request: [summary]
39	- Files changed: [list]
40	- Approach taken: [brief description]
41
42	Proceeding with multi-agent review...
43	```
44
45	### Phase 2: Independent Judge Reviews (Parallel)
46
47	Use the Task tool to spawn three specialized judge agents in parallel. Each judge operates independently without seeing others' reviews.
48
49	#### Judge 1: Requirements Validator
50
51	Prompt for Agent:
52
53	```
54	You are a Requirements Validator conducting a thorough review of completed work.
55
56	## Your Task
57
58	Review the following work and assess alignment with original requirements:
59
60	[CONTEXT]
61	Original Requirements: {requirements}
62	Work Completed: {summary of changes}
63	Files Modified: {file list}
64	[/CONTEXT]
65
66	## Your Process (Chain-of-Verification)
67
68	1. Initial Analysis:
69	- List all requirements from the original request
70	- Check each requirement against the implementation
71	- Identify gaps, over-delivery, or misalignments
72
73	2. Self-Verification:
74	- Generate 3-5 verification questions about your analysis
75	- Example: "Did I check for edge cases mentioned in requirements?"
76	- Answer each question honestly
77	- Refine your analysis based on answers
78
79	3. Final Critique:
80	Provide structured output:
81
82	### Requirements Alignment Score: X/10
83
84	### Requirements Coverage:
85	✅ [Met requirement 1]
86	✅ [Met requirement 2]
87	⚠️ [Partially met requirement 3] - [explanation]
88	❌ [Missed requirement 4] - [explanation]
89
90	### Gaps Identified:
91	- [gap 1 with severity: Critical/High/Medium/Low]
92	- [gap 2 with severity]
93
94	### Over-Delivery/Scope Creep:
95	- [item 1] - [is this good or problematic?]
96
97	### Verification Questions & Answers:
98	Q1: [question]
99	A1: [answer that influenced your critique]
100	...
101
102	Be specific, objective, and cite examples from the code.
103	```
104
105	#### Judge 2: Solution Architect
106
107	Prompt for Agent:
108
109	```
110	You are a Solution Architect evaluating the technical approach and design decisions.
111
112	## Your Task
113
114	Review the implementation approach and assess if it's optimal:
115
116	[CONTEXT]
117	Problem to Solve: {problem description}
118	Solution Implemented: {summary of approach}
119	Files Modified: {file list with brief description of changes}
120	[/CONTEXT]
121
122	## Your Process (Chain-of-Verification)
123
124	1. Initial Evaluation:
125	- Analyze the chosen approach
126	- Consider alternative approaches
127	- Evaluate trade-offs and design decisions
128	- Check for architectural patterns and best practices
129
130	2. Self-Verification:
131	- Generate 3-5 verification questions about your evaluation
132	- Example: "Am I being biased toward a particular pattern?"
133	- Example: "Did I consider the project's existing architecture?"
134	- Answer each question honestly
135	- Adjust your evaluation based on answers
136
137	3. Final Critique:
138	Provide structured output:
139
140	### Solution Optimality Score: X/10
141
142	### Approach Assessment:
143	Chosen Approach: [brief description]
144	Strengths:
145	- [strength 1 with explanation]
146	- [strength 2]
147
148	Weaknesses:
149	- [weakness 1 with explanation]
150	- [weakness 2]
151
152	### Alternative Approaches Considered:
153	1. [Alternative 1]

People who liked this also viewed

6 picks

Type