name	dev-loop
description	Research → Execute → Dual-Review Loop (Code Quality + Issue Resolution) bis beide Reviews bestanden
version	1.0
requires	{"bins":["git"],"env":[],"os":[],"providers":[]}
tags	["dev","review","quality","issue","feature","coding"]
config	{"max_iterations":10,"timeout_minutes":150}

System Prompt Addition

Implement tasks in a structured 3-phase loop:

Research+Plan (merged into ONE subprocess call): Analyze the codebase, read relevant files, AND produce a concrete implementation plan in the same response. Search the web only when local code and docs are insufficient. Output: .dev-loop/research-and-plan.md.
Execute: Implement the solution based on the research+plan output. On subsequent iterations, also fix all issues found in previous reviews.
Dual-Review (both must pass before finishing, both read-only):
- Quality Review (P1/P2/P3): correctness, security, performance, maintainability, testability, robustness, documentation, compliance. P1 = critical/crash/security | P2 = significant | P3 = minor (non-blocking)
- Resolution Review (RESOLVED/PARTIAL/UNRESOLVED): does the code solve the original task 100%? Quality is irrelevant here — only task completion matters. Both review phases re-read git diff fresh; they do not trust pinned context.

Loop repeats until both reviews pass. No auto-commit or push. Output written to .dev-loop/.

When CLAUDE_SESSION_ENABLED=true (opt-in), all phases share a Claude --session-id/--resume session for prompt-cache reuse, with iteration-cap=5 rollover.

Más de este repositorio

mismo repositorio

brainstorm

swDomass/AI_orchestrator

Multi-persona Round-Table mit domain-aware Personas (LLM wählt 4–6 themenspezifische Rollen); iterative Cross-Pollination bis TF-IDF-Konvergenz; Synthesizer ranked Top-N mit Pro/Contra

2026-05-165

deep-security-audit

swDomass/AI_orchestrator

Multi-agent deep security audit — 6 expert personas (pentester, architect, code auditor, supply chain, data privacy, forensics) + CISO synthesis + optional fix implementation

2026-03-275

critical-review

swDomass/AI_orchestrator

Radical-honesty architectural review — questions code, methodology, design, and operational fitness

2026-03-195

security-audit

swDomass/AI_orchestrator

Security audit + fix — scans for hardcoded secrets, injection vectors, path traversal, input validation gaps, then implements all fixes

2026-03-195

knowledge-transfer

swDomass/AI_orchestrator

Cross-Domain Know-How-Transfer — findet kreative Anwendungen von Vault-Expertise in anderen Branchen

2026-03-095

research-qa

swDomass/AI_orchestrator

Deep Research + Fragen-Katalog vor Implementierungsbeginn

2026-03-075

name	dev-loop
description	Research → Execute → Dual-Review Loop (Code Quality + Issue Resolution) bis beide Reviews bestanden
version	1.0
requires	{"bins":["git"],"env":[],"os":[],"providers":[]}
tags	["dev","review","quality","issue","feature","coding"]
config	{"max_iterations":10,"timeout_minutes":150}

System Prompt Addition

Implement tasks in a structured 3-phase loop:

Research+Plan (merged into ONE subprocess call): Analyze the codebase, read relevant files, AND produce a concrete implementation plan in the same response. Search the web only when local code and docs are insufficient. Output: .dev-loop/research-and-plan.md.
Execute: Implement the solution based on the research+plan output. On subsequent iterations, also fix all issues found in previous reviews.
Dual-Review (both must pass before finishing, both read-only):
- Quality Review (P1/P2/P3): correctness, security, performance, maintainability, testability, robustness, documentation, compliance. P1 = critical/crash/security | P2 = significant | P3 = minor (non-blocking)
- Resolution Review (RESOLVED/PARTIAL/UNRESOLVED): does the code solve the original task 100%? Quality is irrelevant here — only task completion matters. Both review phases re-read git diff fresh; they do not trust pinned context.

Loop repeats until both reviews pass. No auto-commit or push. Output written to .dev-loop/.

When CLAUDE_SESSION_ENABLED=true (opt-in), all phases share a Claude --session-id/--resume session for prompt-cache reuse, with iteration-cap=5 rollover.