Coding & Developmentπ€claudeπuniversal
π§ lossless-code
DAG-based Lossless Context Management for Claude Code. Every message preserved, summaries cascade as a DAG, patterns extracted via Lossless Dream β full recall and reflection across sessions.
Quick Start
Install this Skill
npx skills add GodsBoy/lossless-code<div align="center">
# π§ lossless-code
**DAG-based Lossless Context Management for Claude Code.**
*Every message preserved forever. Summaries cascade, never delete. Full recall across sessions.*
[](https://opensource.org/licenses/MIT)
[](https://github.com/GodsBoy/lossless-code/stargazers)
[](https://github.com/GodsBoy/lossless-code/network/members)
[](https://github.com/GodsBoy/lossless-code/issues)
[](https://github.com/GodsBoy/lossless-code/commits/main)
[](https://www.python.org/)
[](https://www.sqlite.org/)
[](https://docs.anthropic.com/en/docs/claude-code)
[](https://modelcontextprotocol.io/)
[](https://github.com/GodsBoy/lossless-code/pulls)
[Getting Started](#install) Β· [MCP Server](#mcp-server) Β· [Commands](#commands) Β· [Dream](#lossless-dream) Β· [Terminal UI](#terminal-ui-lcc-tui) Β· [How It Works](#how-it-works) Β· [Configuration](#configuration) Β· [Contributing](#contributing)
</div>
---
> Claude Code forgets. claude-mem remembers fragments. lossless-code remembers everything.
## Try it in 60 seconds
```
/plugin marketplace add GodsBoy/lossless-code
/plugin install lossless-code
```
That's it. Start a new session and search your history:
> lcc_grep "database migration"
## The Problem
Claude Code forgets everything between sessions. Memory tools like ClawMem, context-memory, and claude-mem use flat retrieval: keyword search over snippets, no structure, no hierarchy, no way to trace a summary back to its source conversation.
When a project spans weeks and hundreds of sessions, flat search fails. You get fragments without lineage.
## What Makes lossless-code Different
lossless-code uses **DAG-based lossless preservation**, the same approach pioneered by [lossless-claw](https://github.com/Martian-Engineering/lossless-claw) for [OpenClaw](https://github.com/openclaw/openclaw):
- **Nothing is ever deleted.** Every message stays in `vault.db` forever.
- **Summaries form a directed acyclic graph.** Messages cascade to depth-0 summaries, which roll up to depth-1, depth-2, and beyond.
- **Full drill-down.** `lcc_expand` traces any summary back to the exact messages that created it.
- **Automatic.** Claude Code hooks capture every turn and trigger summarisation transparently. Zero manual effort.
- **Cross-session recall.** Start a new session and your full project history is immediately searchable.
- **Lossless Dream.** Extracts recurring patterns (corrections, preferences, conventions) from vault history and injects them into future sessions β like Auto-Dream but without forgetting.
```
ββββββββββββββββββββ
β Claude Code β
β Session β
ββββββββββ¬ββββββββββ
β
ββββββββββββββββββββββββββΌβββββββββββββββββββββββββ
β β β β
βββββββΌββββββ ββββββββΌβββββββ ββββββββΌβββββββ ββββββΌββββββ
β Hooks β β Skills β β CLI β β MCP β
β (write) β β (shell) β β Tools β β Server β
β β β β β β β (stdio) β
β SessionStartβ β lcc_grep β β lcc_status β β β
β Stop β β lcc_expand β β β β 6 tools β
β PreCompact β β lcc_context β β β β read-onlyβ
β PostCompactβ β lcc_sessionsβ β β β β
β UserPrompt β β lcc_handoff β β β β β
βββββββ¬βββββββ ββββββββ¬βββββββ ββββββββ¬βββββββ ββββββ¬ββββββ
β β β β
ββββββββββββββββββββΌβββββββββββββββββΌβββββββββββββββ
β β
ββββββββββΌβββββββββββββββββΌβββ
β vault.db β
β
... (truncated)Repository
4stars
1forks
Details
Related Skills
More coding & development skills you might like
Reticle
Coding & DevelopmentAI agents can generate code, but still struggle to understand what they build. Reticle brings Jev-style machine-native runtime perception to web & desktop applications.
π€claudeπuniversal
1.2k
190
View
agentscope-java
Coding & DevelopmentExpert Java developer skill for AgentScope Java framework - a reactive, message-driven multi-agent system built on Project Reactor. Use when working with reactive programming, LLM integration, agent orchestration, multi-agent systems, or when the user mentions AgentScope, ReActAgent, Mono/Flux, Project Reactor, or Java agent development. Specializes in non-blocking code, tool integration, hooks, pipelines, and production-ready agent applications.
πuniversal
6.0k
1.5k
View
ospec
Coding & DevelopmentDocument-driven OSpec workflow for initialization, change/goal routing, validation, archiving, and durable project knowledge.
β¨geminiπ€claudeπ¬gptπuniversal
524
34
View
misakanet-failure-memory
Coding & Developmentπ A zero-dependency, git-backed micro-lesson library for AI Agents to asynchronously share and search verified debugging experience. Python stdlib only. | https://misakanet.org
π€claudeβ¨οΈcursorπuniversal
424
158
View