parcadei

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

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

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

Problem-solving strategies for constrained optimization.

Problem-solving strategies for convex optimization.

Problem-solving strategies for gradient methods in optimization.

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

Planning agent that creates implementation plans and handoffs from conversation context.

Automatically trace Claude Code conversations to Braintrust for observability, capturing sessions, conversation turns, and tool calls as hierarchical traces.

This skill ensures an agent's operational context is isolated, preventing interference from other agents or external data.

Agentic Workflow Pattern.

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

Meta-skill for internal codebase exploration at varying depths (quick, deep, or architectural).

Problem-solving strategies for source coding in information theory.

Problem-solving strategies for vector spaces in linear algebra.

Problem-solving strategies for limits in real analysis.

Problem-solving strategies for compactness in topology.

Search past reasoning for relevant decisions and approaches.

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

Wiring Verification

Guide for integrating Agentica SDK with Claude Code CLI proxy.

Reference guide for Agentica multi-agent infrastructure APIs.

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

Build Python agents with the Agentica SDK, featuring the @agentic decorator, spawn(), persistence, and MCP integration.

Agentica server and Claude proxy setup: architecture, startup sequence, and debugging.

Async REPL Protocol

Search Tool Hierarchy

Search Tool Hierarchy

Async REPL Protocol

Analyze Claude Code sessions via Braintrust

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

Create git commits with user approval and no Claude attribution.

A complete skill for E2E testing.

Completion Check: Verify Infrastructure Is Wired.

Transforms session learnings into permanent capabilities such as skills, rules, and agents. Use this when asked to improve setup, learn from sessions, compound learnings, or identify patterns that should become skills.

Creates a handoff document for transferring work to another session.

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

Generate comprehensive PR descriptions following repository templates.

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

Environment Triage

Graceful Degradation with Helpful Messages.

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

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

Idempotent redundancy ensures that repeating an operation does not cause additional side effects, maintaining the same final state.

Implement technical plans from thoughts, shared ideas, or existing plans, with verification.

Implement technical plans from thoughts/shared/plans with verification.

Search Mathlib for lemmas by type signature pattern.

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