parcadei

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

Wiring Verification

Validation agent that validates plan tech choices against current best practices

Friendly onboarding when users ask about capabilities

Show full session token usage, costs, TLDR savings, and hook activity

Maps questions to the optimal tldr command. Use this to pick the right layer

Get a token-efficient overview of any project using the TLDR stack

Full 5-layer analysis of a specific function. Use when debugging or deeply understanding code.

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

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

Orchestrator-only workflow for migrating/rewriting codebases with full TDD and agent delegation

TDD workflow for migrations - orchestrate agents, zero main context growth

Test-driven development workflow with philosophy guide - plan → write tests → implement → validate

Show users how Continuous Claude works - the opinionated setup with hooks, memory, and coordination

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

Create and use Claude Code slash commands - quick prompts, bash execution, file references

Upgrade any skill to v5 Hybrid format using decision theory + modal logic

Skill Development Rules

Meta-skill for creating and managing Claude Code skills

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

Security audit workflow - vulnerability scan → verification

Search Tool Hierarchy

Choose the right search tool for each query type

Search Tool Hierarchy

Router-First Architecture

Comprehensive code review workflow - parallel specialized reviews → synthesis

Resume work from handoff document with context analysis and validation

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

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

Document codebase as-is with thoughts directory for historical context

Use RepoPrompt CLI for token-efficient codebase exploration

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

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

Release preparation workflow - security audit → E2E tests → review → changelog → docs

Check reference SDK implementations using btca ask

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

Search past reasoning for relevant decisions and approaches

Query the memory system for relevant learnings from past sessions

QLTY During Development

Code quality checks, formatting, and metrics via qlty CLI

Formal theorem proving with research, testing, and verification phases

Identify failure modes before they occur using structured risk analysis

Planning agent that creates implementation plans and handoffs from conversation context

Unit-aware computation with Pint - convert units, dimensional analysis, unit arithmetic

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

Parallel Agent Orchestration

Parallel Agent Type Contracts

OPC Architecture Understanding

Analyze brownfield codebase and create initial continuity ledger

Observe Before Editing

Never Use TaskOutput

No Polling for Background Agents

Search library documentation and code examples via Nia

System health check (MOT) for skills, agents, hooks, and memory

Fast codebase search via WarpGrep (20x faster than grep)

Fast file editing via Morph Apply API (10,500 tokens/sec, 98% accuracy)

Modular Code Organization

Migration workflow - research → analyze → plan → implement → review

MCP Script Rules

Research-to-implement pipeline chaining 5 MCP tools with graceful degradation

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

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

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

Search Mathlib for lemmas by type signature pattern

LLM Tuning Patterns

Index at Creation Time

Implementation agent that executes a single task and creates handoff on completion

Implement technical plans from thoughts/shared/plans with verification

Implement technical plans from thoughts/shared/plans with verification

Idempotent Redundancy

Hook Development Rules

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

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

Graceful Degradation with Helpful Messages

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

Git Commit Rules

Meta-skill workflow orchestrator for bug investigation and resolution. Routes to debug, implement, test, and commit based on scope.

Scrape web pages and extract content via Firecrawl MCP

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

Explicit Identity Across Boundaries

Environment Triage

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

Generate comprehensive PR descriptions following repository templates

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

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

Find unused functions and dead code in the codebase

Create handoff document for transferring work to another session

Create or update continuity ledger for state preservation across clears

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".

Completion Check: Verify Infrastructure Is Wired

A complete skill for E2E testing

Create git commits with user approval and no Claude attribution

Claude Code CLI commands, flags, headless mode, and automation patterns

Workflow orchestrator that chains existing skills for feature development

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

Analyze Claude Code sessions via Braintrust

Background Agent Pings

Async REPL Protocol

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

Spawn Agentica multi-agent patterns

Agentica server + Claude proxy setup - architecture, startup sequence, debugging

Build Python agents with Agentica SDK - @agentic decorator, spawn(), persistence, MCP integration

Write reliable prompts for Agentica/REPL agents that avoid LLM instruction ambiguity

Reference guide for Agentica multi-agent infrastructure APIs

Guide for integrating Agentica SDK with Claude Code CLI proxy

Agentic Workflow Pattern

Agent Orchestration Rules

Agent Context Isolation