continuous-claude-v3

Deep interview process to transform vague ideas into detailed specs. Works for technical and non-technical users.

Identify failure modes before they occur using structured risk analysis

Unified math capabilities - computation, solving, and explanation. I route to the right tool.

Search GitHub code, repositories, issues, and PRs via MCP

Find unused functions and dead code in the codebase

Code refactoring workflow - analyze → plan → implement → review → validate

Search past reasoning for relevant decisions and approaches

Graceful Degradation with Helpful Messages

Meta-skill for internal codebase exploration at varying depths (quick/deep/architecture)

Agentic Workflow Pattern

Modular Code Organization

Guide to the math cognitive stack - what tools exist and when to use each

Create and configure Claude Code sub-agents with custom prompts, tools, and models

Security audit workflow - vulnerability scan → verification

Research agent for external documentation, best practices, and library APIs via MCP tools

AI-powered web search, research, and reasoning via Perplexity

Agent Orchestration Rules

Store a learning, pattern, or decision in the memory system for future recall

Git Commit Rules

Analyze repository structure, patterns, conventions, and documentation for understanding a new codebase

Comprehensive code review workflow - parallel specialized reviews → synthesis

AST-based code search and refactoring via ast-grep MCP

Planning agent that creates implementation plans and handoffs from conversation context

Use RepoPrompt CLI for token-efficient codebase exploration

Agent Context Isolation

Parallel Agent Orchestration

Create or update continuity ledger for state preservation across clears

Create handoff document for transferring work to another session

Debug issues by investigating logs, database state, and git history

Hook Development Rules

Formal theorem proving with research, testing, and verification phases

Transform session learnings into permanent capabilities (skills, rules, agents). Use when asked to "improve setup", "learn from sessions", "compound learnings", or "what patterns should become skills".

Complete Claude Code hooks reference - input/output schemas, registration, testing patterns

Token-efficient code analysis via 5-layer stack (AST, Call Graph, CFG, DFG, PDG). 95% token savings.

External research workflow for docs, web, APIs - NOT codebase exploration

Comprehensive testing workflow - unit tests ∥ integration tests → E2E tests

Background Agent Pings

Create git commits with user approval and no Claude attribution

Systematic hook debugging workflow. Use when hooks aren't firing, producing wrong output, or behaving unexpectedly.

Generate comprehensive PR descriptions following repository templates

Implement technical plans from thoughts/shared/plans with verification

Braintrust tracing for Claude Code - hook architecture, sub-agent correlation, debugging

OPC Architecture Understanding

Computational geometry with Shapely - create geometries, boolean operations, measurements, predicates

Goal-based workflow orchestration - routes tasks to specialist agents based on user goals

Deterministic router for math cognitive stack - maps user intent to exact CLI commands

Query the memory system for relevant learnings from past sessions

MCP Script Rules

Interactive workspace discovery - learn what tools, workflows, agents, and hooks are available

Scrape web pages and extract content via Firecrawl MCP