do-competitively

Name: do-competitively
Author: NeoLabHQ

852installs132forks

How to install do-competitively?

neolabhq/context-engineering-kit/do-competitively

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

Installs into the current project.

Summary

Execute tasks through competitive multi-agent generation, meta-judge evaluation specification, multi-judge evaluation, and evidence-based synthesis

Files · 1

View on GitHub

SKILL.md

1	# do-competitively
2
3	<task>
4	Execute tasks through competitive multi-agent generation, meta-judge evaluation specification, multi-judge evaluation, and evidence-based synthesis to produce superior results by combining the best elements from parallel implementations.
5	</task>
6
7	<context>
8	This command implements the Generate-Critique-Synthesize (GCS) pattern with adaptive strategy selection for high-stakes tasks where quality matters more than speed. It combines competitive generation with meta-judge evaluation specification and multi-perspective evaluation, then intelligently selects the optimal synthesis strategy based on results.
9
10	Key features:
11
12	- Self-critique loops in generation (Constitutional AI)
13	- Structured evaluation - Meta-judge produces tailored rubrics before judging
14	- Verification loops in evaluation (Chain-of-Verification)
15	- Adaptive strategy: polish clear winners, synthesize split decisions, redesign failures
16	- Average 15-20% cost savings through intelligent strategy selection
17	</context>
18
19	CRITICAL: You are not implementation agent or judge, you shoudn't read files that provided as context for sub-agent or task. You shouldn't read reports, you shouldn't overwhelm your context with unneccesary information. You MUST follow process step by step. Any diviations will be considered as failure and you will be killed!
20
21	## Pattern: Generate-Critique-Synthesize (GCS)
22
23	This command implements a multi-phase adaptive competitive orchestration pattern:
24
25	```
26	Phase 1: Competitive Generation with Self-Critique + Meta-Judge (IN PARALLEL)
27	┌─ Meta-Judge → Evaluation Specification YAML ───────────┐
28	Task ────┼─ Agent 2 → Draft → Critique → Revise → Solution B ───┐ │
29	├─ Agent 3 → Draft → Critique → Revise → Solution C ───┼─┤
30	└─ Agent 1 → Draft → Critique → Revise → Solution A ───┘ │
31	│
32	Phase 2: Multi-Judge Evaluation with Verification │
33	┌─ Judge 1 → Evaluate → Verify → Revise → Report A ─┐ │
34	├─ Judge 2 → Evaluate → Verify → Revise → Report B ─┼────┤
35	└─ Judge 3 → Evaluate → Verify → Revise → Report C ─┘ │
36	│
37	Phase 2.5: Adaptive Strategy Selection │
38	Analyze Consensus ───────────────────────────────────────┤
39	├─ Clear Winner? → SELECT_AND_POLISH │
40	├─ All Flawed (<3.0)? → REDESIGN (return Phase 1) │
41	└─ Split Decision? → FULL_SYNTHESIS │
42	│ │
43	Phase 3: Evidence-Based Synthesis │ │
44	(Only if FULL_SYNTHESIS) │ │
45	Synthesizer ─────────────────────┴───────────────────────┴─→ Final Solution
46	```
47
48	## Process
49
50	### Setup: Create Reports Directory
51
52	Before starting, ensure the reports directory exists:
53
54	```bash
55	mkdir -p .specs/reports
56	```
57
58	Report naming convention: `.specs/reports/{solution-name}-{YYYY-MM-DD}.[1\|2\|3].md`
59
60	Where:
61
62	- `{solution-name}` - Derived from output path (e.g., `users-api` from output `specs/api/users.md`)
63	- `{YYYY-MM-DD}` - Current date
64	- `[1\|2\|3]` - Judge number
65
66	Note: Solutions remain in their specified output locations; only evaluation reports go to `.specs/reports/`
67
68	### Phase 1: Competitive Generation + Meta-Judge (IN PARALLEL)
69
70	Launch 3 independent generator agents AND 1 meta-judge agent in parallel (4 agents total, all recommended: Opus for quality):
71
72	The meta-judge runs in parallel with the 3 generators because it does not need their output — it only needs the task description to generate evaluation criteria.
73
74	CRITICAL: Dispatch all 4 agents in a single message using 4 Task tool calls as foreground agents. The meta-judge MUST be the first tool call in the dispatch order, because he should have time to collect context from codebase, before it was modified by generators.
75
76	#### Meta-Judge Agent (1 agent)
77
78	The meta-judge generates an evaluation specification YAML (rubrics, checklists, scoring criteria) tailored to this specific task. It returns the evaluation specification YAML that all 3 judges will use.
79
80	Prompt template for meta-judge:
81
82	```markdown
83	## Task
84
85	Generate an evaluation specification yaml for the following task. You will produce rubrics, checklists, and scoring criteria that judge agents will use to evaluate and compare competitive implementation artifacts.
86
87	CLAUDE_PLUGIN_ROOT=`${CLAUDE_PLUGIN_ROOT}`
88
89	## User Prompt
90	{Original task description from user}
91
92	## Context
93	{Any relevant codebase context, file paths, constraints}
94
95	## Artifact Type
96	{code \| documentation \| configurati

People who liked this also viewed

6 picks

Type

tddSkills for Real Engineers. Straight from my .agents directory.SkillsJul 2026501k181k
test-driven-developmentUse when implementing any feature or bugfix, before writing implementation codeSkillsJul 2026175k259k
qaSkills for Real Engineers. Straight from my .agents directory.SkillsJul 2026160k181k
verification-before-completionUse when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidenceSkillsJul 2026153k259k
webapp-testingToolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser lSkillsJul 2026120k163k
playwright-best-practicesUse when writing Playwright tests, fixing flaky tests, debugging failures, implementing Page Object Model, configuring CI/CD, optimizing performance, mocking APIs, handling authentication or OAuth, teSkillsJul 202664k338

do-competitively

852installs132forks

View on GitHub

How to install do-competitively?

neolabhq/context-engineering-kit/do-competitively

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

Installs into the current project.

Summary

Execute tasks through competitive multi-agent generation, meta-judge evaluation specification, multi-judge evaluation, and evidence-based synthesis

Files · 1

View on GitHub

SKILL.md

1	# do-competitively
2
3	<task>
4	Execute tasks through competitive multi-agent generation, meta-judge evaluation specification, multi-judge evaluation, and evidence-based synthesis to produce superior results by combining the best elements from parallel implementations.
5	</task>
6
7	<context>
8	This command implements the Generate-Critique-Synthesize (GCS) pattern with adaptive strategy selection for high-stakes tasks where quality matters more than speed. It combines competitive generation with meta-judge evaluation specification and multi-perspective evaluation, then intelligently selects the optimal synthesis strategy based on results.
9
10	Key features:
11
12	- Self-critique loops in generation (Constitutional AI)
13	- Structured evaluation - Meta-judge produces tailored rubrics before judging
14	- Verification loops in evaluation (Chain-of-Verification)
15	- Adaptive strategy: polish clear winners, synthesize split decisions, redesign failures
16	- Average 15-20% cost savings through intelligent strategy selection
17	</context>
18
19	CRITICAL: You are not implementation agent or judge, you shoudn't read files that provided as context for sub-agent or task. You shouldn't read reports, you shouldn't overwhelm your context with unneccesary information. You MUST follow process step by step. Any diviations will be considered as failure and you will be killed!
20
21	## Pattern: Generate-Critique-Synthesize (GCS)
22
23	This command implements a multi-phase adaptive competitive orchestration pattern:
24
25	```
26	Phase 1: Competitive Generation with Self-Critique + Meta-Judge (IN PARALLEL)
27	┌─ Meta-Judge → Evaluation Specification YAML ───────────┐
28	Task ────┼─ Agent 2 → Draft → Critique → Revise → Solution B ───┐ │
29	├─ Agent 3 → Draft → Critique → Revise → Solution C ───┼─┤
30	└─ Agent 1 → Draft → Critique → Revise → Solution A ───┘ │
31	│
32	Phase 2: Multi-Judge Evaluation with Verification │
33	┌─ Judge 1 → Evaluate → Verify → Revise → Report A ─┐ │
34	├─ Judge 2 → Evaluate → Verify → Revise → Report B ─┼────┤
35	└─ Judge 3 → Evaluate → Verify → Revise → Report C ─┘ │
36	│
37	Phase 2.5: Adaptive Strategy Selection │
38	Analyze Consensus ───────────────────────────────────────┤
39	├─ Clear Winner? → SELECT_AND_POLISH │
40	├─ All Flawed (<3.0)? → REDESIGN (return Phase 1) │
41	└─ Split Decision? → FULL_SYNTHESIS │
42	│ │
43	Phase 3: Evidence-Based Synthesis │ │
44	(Only if FULL_SYNTHESIS) │ │
45	Synthesizer ─────────────────────┴───────────────────────┴─→ Final Solution
46	```
47
48	## Process
49
50	### Setup: Create Reports Directory
51
52	Before starting, ensure the reports directory exists:
53
54	```bash
55	mkdir -p .specs/reports
56	```
57
58	Report naming convention: `.specs/reports/{solution-name}-{YYYY-MM-DD}.[1\|2\|3].md`
59
60	Where:
61
62	- `{solution-name}` - Derived from output path (e.g., `users-api` from output `specs/api/users.md`)
63	- `{YYYY-MM-DD}` - Current date
64	- `[1\|2\|3]` - Judge number
65
66	Note: Solutions remain in their specified output locations; only evaluation reports go to `.specs/reports/`
67
68	### Phase 1: Competitive Generation + Meta-Judge (IN PARALLEL)
69
70	Launch 3 independent generator agents AND 1 meta-judge agent in parallel (4 agents total, all recommended: Opus for quality):
71
72	The meta-judge runs in parallel with the 3 generators because it does not need their output — it only needs the task description to generate evaluation criteria.
73
74	CRITICAL: Dispatch all 4 agents in a single message using 4 Task tool calls as foreground agents. The meta-judge MUST be the first tool call in the dispatch order, because he should have time to collect context from codebase, before it was modified by generators.
75
76	#### Meta-Judge Agent (1 agent)
77
78	The meta-judge generates an evaluation specification YAML (rubrics, checklists, scoring criteria) tailored to this specific task. It returns the evaluation specification YAML that all 3 judges will use.
79
80	Prompt template for meta-judge:
81
82	```markdown
83	## Task
84
85	Generate an evaluation specification yaml for the following task. You will produce rubrics, checklists, and scoring criteria that judge agents will use to evaluate and compare competitive implementation artifacts.
86
87	CLAUDE_PLUGIN_ROOT=`${CLAUDE_PLUGIN_ROOT}`
88
89	## User Prompt
90	{Original task description from user}
91
92	## Context
93	{Any relevant codebase context, file paths, constraints}
94
95	## Artifact Type
96	{code \| documentation \| configurati

People who liked this also viewed

6 picks

Type