# EGC: Your AI agents never start from zero again

> Source: <https://dev.to/fmarzochi/egc-your-ai-agents-never-start-from-zero-again-1o09>
> Published: 2026-06-20 21:55:37+00:00

Every time you open a new session with an AI coding tool, it starts from zero. It does not know what you decided yesterday, what failed last week, or what comes next. You have to explain the project again. And again.

**EGC (Extended Global Context)** fixes this.

EGC is a local runtime that gives every AI coding tool you use a persistent memory. At the end of each session, the AI saves what it learned: decisions made, what failed, your preferences, what comes next. At the start of the next session, it loads that state back automatically.

One install. Every tool. Every session.

Website: [https://fmarzochi.github.io/EGCSite](https://fmarzochi.github.io/EGCSite)

You open Claude Code on a project you have not touched in two weeks. Without typing anything:

```
State loaded from egc-memory via ~/.egc/state/Projects--MyApp.md
Context and preferences acknowledged.
Ready to pick up:
  - Test full install on a clean machine
  - Add GEMINI.md with session memory protocol
  - Publish v1.0.1 fix after clean install test passes
```

The AI already knows what you were building, what decisions you made, what failed, and exactly where you stopped. You did not type anything. You just started working.

EGC ships two MCP servers that run locally during every session.

| Tool | What it does |
|---|---|
`get_state` |
Loads project memory at session start |
`update_state` |
Saves decisions, preferences, and next steps |
`store_decision` |
Persists a single decision to SQLite |
`query_history` |
Returns past decisions by timestamp |
`search_history` |
Full-text search with BM25 ranking |
`working_memory_set` |
Stores transient context with a TTL |
`lesson_save` |
Records cross-session knowledge with confidence decay |
`lesson_recall` |
Retrieves active lessons above a threshold |
`detect_patterns` |
Surfaces repeated commands and recurring errors |
`compress_observations` |
Compresses hook events to save token budget |

State files live at `~/.egc/state/<project-slug>.md`

. One file per project. Plain Markdown. Human-readable.

| Tool | What it does |
|---|---|
`validate_command` |
Checks shell commands before execution |
`validate_write` |
Validates file paths to prevent unsafe writes |
`reduce_context` |
Compresses file payloads to save token budget |
`orchestrate_task` |
Routes prompts with agent/skill context |
`auto_learn` |
Mines session failures and writes lessons to CLAUDE.md |

EGC installs into whichever tools you already use:

All of them read from and write to the same `~/.egc/state/`

directory. What Claude knows, Gemini knows. No copy-paste, no re-explanation.

```
npm install -g @egchq/egc && egc install
```

Or without installing globally:

```
npx @egchq/egc install
```

EGC detects which tools you have installed and sets them up automatically.

EGC ships 479 components as a bonus: 63 agents, 229 skills, and 76 commands written from real engineering sessions. Skip them entirely and EGC still gives you persistent memory. Or use them to extend any AI tool you already have.

```
npx skills add Fmarzochi/EGC
```

EGC is MIT licensed, built by one developer, maintained in the open.

`@egchq/egc`

If EGC changed how you work, a star goes a long way.