parcadeiCommunity·144 Agent Skills Included

Continuous-Claude-v3

Self-improving Claude Code setup with memory, agents, and workflows

Turns Claude Code into a persistent, learning development environment with 109 skills, 32 agents, and 30 hooks. Eliminates context loss between sessions, wasted tokens on file reads, and repeated manual workflows. Adds semantic memory recall, token-efficient code analysis, multi-agent orchestration, and math/proof tooling.

3,934|299

↗

npx skills add parcadei/Continuous-Claude-v3 --all -g -y

All Skills in This Repository (144)

Pure Emerald Level Indicators

📦 In Repo

parcadei

research

Document existing codebase components and interactions into structured Markdown reports.

commit

Analyze code changes, plan logical commit groupings, and draft imperative commit messages.

index-at-creation

Index artifacts immediately upon creation using event-driven triggers.

tldr-stats

Generate a statistical dashboard of Claude API token usage, costs, and TLDR savings.

workflow-router

Route tasks to specialist agents based on user goals and resource preferences.

graceful-degradation

Check optional service availability and provide actionable fallback messages.

async-repl-protocol

Enforce await syntax and single code blocks in Agentica async REPL tests.

describe-pr

Generate pull request descriptions from code diffs, commits, and templates.

skill-development

Standardize SKILL.md structure and MCP script integration for skill development.

search-router

Route search queries to the most token-efficient tool by query type.

modular-code

Organize Python code into modular files and packages.

opc-architecture

Explain the architecture and operational flow of Orchestrated Parallel Claude.

Community

Intermediate

Frequently Asked Questions

FAQPage Schema

How to install Continuous-Claude-v3?▼

Run `npx skills add parcadei/Continuous-Claude-v3 --all -g -y` in your terminal to install all skills in this suite globally.

What problem does Continuous Claude solve?▼

It fixes Claude Code's context compaction problem by storing learnings in a database and using YAML handoffs, so each session starts with full context instead of losing prior decisions.

How does Continuous Claude save tokens?▼

Its TLDR code analysis compresses file reads by about 95% using AST, call graph, and data flow summaries instead of reading entire raw files.

Do I need to memorize slash commands?▼

No. You describe your goal in plain English and a hook automatically suggests the right skills and agents, such as /fix for bugs or /build for features.

Does Continuous Claude work without paid services?▼

Yes. The core system runs locally with Docker and PostgreSQL; optional services like Perplexity or Braintrust only add extra research and tracing features.