skills/proffesor-for-testing__agentic-qe

mirror of https://github.com/proffesor-for-testing/agentic-qe.git synced 2026-09-19 08:45:47 +08:00

Files

T

Profa 2253d05618 feat: flatten project structure — archive v2, promote v3 to root

Remove the nested v3/ subdirectory by moving all active source code,
tests, configs, and assets directly to the project root. Archive v2/
(dead code, 847MB) to v2-archive.tar.gz.

Key changes:
- Move v3/src/ → src/, v3/tests/ → tests/, v3/packages/ → packages/
- Move v3/tsconfig.json, v3/vitest.config.ts to root
- Merge v3/package.json into root package.json (ESM, scripts, deps, exports)
- Merge v3/scripts/ into scripts/, v3/implementation/ into docs/implementation/
- Update all CI workflows to remove "cd v3 &&" and v3/ path prefixes
- Fix all init installer path resolution (3-level-up → 2-level-up)
- Fix hooks to use "node ./dist/cli/bundle.js" instead of bare "aqe"
- Update settings.json ADR/DDD directories to docs/implementation/
- Fix helper scripts (statusline, adr-compliance, ddd-tracker, guidance-hooks)
- Update 58 agent definitions: aqe/v3/ → aqe/ memory namespaces
- Update 15 guidance shards: v3/src/ → src/ path references
- Fix sync interfaces: remove v3/ and v2/ path references
- Fix test path calculations (process.cwd() + '..' no longer needed)
- Fix flaky perf test threshold (500ms → 1000ms for CI environments)
- Fix complexity-analyzer test to match ADR-051 keyword-based eligibility

Build passes, 17,901 tests pass, CLI and MCP bundles verified.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

2026-03-01 11:35:41 +00:00

5.1 KiB

Raw Permalink Blame History

SPEC-040-B: RL Algorithm Distribution

Field	Value
Specification ID	SPEC-040-B
Parent ADR	ADR-040
Version	1.0
Status	Implemented
Last Updated	2026-01-13
Author	Claude Code

Overview

Defines the distribution of 8 RL algorithms across 12 DDD domains, including QE-specific applications and integration components.

Specification Details

Section 1: Domain Integration Status (12/12 Complete)

Domain	RL Algorithms	Integration Components	Tests
learning-optimization	QESONA, QEFlashAttention	Coordinator, learning coordinator	104
defect-intelligence	QEFlashAttention	Pattern learner, predictor	87
coverage-analysis	Q-Learning, QEGNNEmbeddingIndex	HNSW index, gap detection	49 + 10 (QL)
test-execution	DecisionTransformer, QESONA	Parallel executor, retry logic, prioritizer	121 + 13 (DT)
test-generation	QESONA, QEFlashAttention, DecisionTransformer	Test case generator	28
quality-assessment	ActorCritic, QESONA, QEFlashAttention	Quality gate, assessment	67
requirements-validation	PPO, QESONA	BDD scenario generator	45
code-intelligence	QEGNNEmbeddingIndex, QESONA	Knowledge graph, semantic search	89
security-compliance	DQN, QEFlashAttention	SAST/DAST scanner	38
contract-testing	SARSA, QESONA	API contract validation, prioritization	17
visual-accessibility	A2C, QEFlashAttention	WCAG compliance, image similarity	21
chaos-resilience	PolicyGradient, QESONA	Fault injection, resilience patterns	25

Section 2: RL Algorithm Distribution

Q-Learning           -> coverage-analysis
Decision Transformer -> test-execution, test-generation
PPO                  -> requirements-validation
Actor-Critic         -> quality-assessment
DQN                  -> security-compliance
SARSA                -> contract-testing
A2C                  -> visual-accessibility
Policy Gradient      -> chaos-resilience
QESONA               -> 11 domains (learning optimization layer)
QEFlashAttention     -> 7 domains (attention optimization)
QEGNNEmbeddingIndex  -> 2 domains (code intelligence, coverage)

Section 3: QE-Specific RL Applications

Algorithm	QE Application	Domain	Status
Decision Transformer	Test case prioritization	test-execution	Implemented
Q-Learning	Coverage path optimization	coverage-analysis	Implemented
SARSA	API contract sequencing	contract-testing	Implemented
Actor-Critic	Quality gate threshold tuning	quality-assessment	Implemented
Policy Gradient	Fault injection strategies	chaos-resilience	Implemented
DQN	Security threat prioritization	security-compliance	Implemented
PPO	Requirements testability scoring	requirements-validation	Implemented
A2C	Visual accessibility scoring	visual-accessibility	Implemented

Section 4: Flash Attention Configuration

const QE_FLASH_ATTENTION_CONFIG = {
  // Memory-efficient attention for large test suites
  blockSize: 64,
  numBlocks: 128,

  // QE workload patterns
  patterns: {
    testSimilarity: {
      headsPerBlock: 8,
      queryChunkSize: 512
    },
    codeEmbedding: {
      headsPerBlock: 4,
      queryChunkSize: 1024
    },
    defectMatching: {
      headsPerBlock: 12,
      queryChunkSize: 256
    }
  }
};

Section 5: Legacy QE API (Preserved)

// Import from QE integrations (backward compatible)
import { SONA, createSONA } from './integrations/rl-suite';
import { FlashAttention } from './integrations/flash-attention';
import { QLearning } from './integrations/rl-suite/algorithms';

// Use SONA for pattern adaptation
const sona = createSONA({ dimension: 384 });
const result = await sona.adaptPattern(state, 'test-generation', 'test-generation');

// Use Flash Attention for similarity computation
const fa = new FlashAttention({ backend: 'wasm-simd' });
const similarity = await fa.computeSimilarity(queryEmbedding, patterns);

Validation Rules

Rule ID	Description	Severity
SPEC-040-B-001	Each domain must have at least one RL algorithm	Error
SPEC-040-B-002	QESONA must be available in all applicable domains	Warning
SPEC-040-B-003	All 8 RL algorithms must be connected	Error

Spec ID	Title	Relationship
SPEC-040-A	Ruvector Wrappers	Wrapper implementations
SPEC-040-C	Performance Assessment	Performance metrics

Changelog

Version	Date	Author	Changes
1.0	2026-01-13	Claude Code	Initial specification

References

Parent ADR
12 DDD domain architecture documentation

5.1 KiB Raw Permalink Blame History