caveman
01Skill Claude Code
Optional concise-response mode. Compresses prose 40-65% while keeping code, paths, identifiers, and security text verbatim.
Copilot agent delegation yapısının proje reposudur.
This repository also configures its own agents. See what agent-delegation tells them →
Skill Claude Code
Optional concise-response mode. Compresses prose 40-65% while keeping code, paths, identifiers, and security text verbatim.
Skill Claude Code
Context window optimizer. Use ctx MCP tools to run analysis, index knowledge, and search prior findings without polluting context.
Skill Claude Code
Build and query knowledge graphs from codebases. Scoped to T4/T5 analysis tiers. Use for codebases >20 files or when codebase topology is unknown.
Skill Claude Code
Generate test suites from implementation files. Supports Jest, Vitest, and pytest. Produces AAA-structured tests covering happy path, error paths, and edge cases.
Skill Codex
Deep analysis and research skill for codebase exploration, document reading, dependency mapping, risk assessment, hotspot discovery, and test scenario generation.
Skill Codex
Integration skill for external/internal API consumption, contract alignment, mapping, error handling, and reliability concerns around network boundaries.
Skill Codex
Architecture decision skill for ADR-oriented tradeoff analysis, module boundary decisions, and pattern selection work.
Skill Codex
Backend implementation skill covering service boundaries, contracts, validation, failure handling, and side-effect control.
Skill Codex
Security baseline skill for untrusted input handling, auth/session concerns, secrets hygiene, and common backend exploit classes.
Skill Codex
Comprehensive code hygiene skill. Defines naming, decomposition, readability, SOLID alignment, prohibition rules, and review heuristics for maintainable production code.
Skill Codex
Architecture skill covering module boundaries, layering, dependency direction, tradeoff analysis, and ADR-grade decision framing.
Skill Codex
Review skill for severity-based evaluation of correctness, security, architecture alignment, maintainability, and testing sufficiency.
Skill Codex
Conventional commit and atomic change discipline skill.
Skill Codex
Deep token-efficiency skill covering dynamic context loading, query-first analysis, progressive disclosure, and context-budget enforcement.
Skill Codex
Frontend development skill covering JavaScript/TypeScript, React version-aware development, UI state modeling, component architecture, forms, accessibility, and design-system compliance.
Skill Codex
Delivery-oriented skill for turning requirements into bounded, testable, production-appropriate changes.
Skill Codex
JVM-focused quality gate skill covering static analysis, injection discipline, test/coverage compatibility, and common code-quality tooling expectations.
Skill Codex
Graph-assisted codebase navigation skill for large repositories, hotspot detection, coupling analysis, and topology-first discovery.
Skill Codex
Pull request quality skill for scope control, reviewer usability, deployment-risk communication, and changelog-quality summaries.
Skill Codex
Quick security audit skill for scanning common high-risk patterns and producing triage-oriented findings.
Skill Codex
Test design skill for validating behavior changes with happy-path, edge, failure, and regression coverage.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: