Start NanoClaw v2 — ECC's persistent, zero-dependency REPL with model routing, skill hot-load, branching, compaction, export, and metrics.

Analyze context window usage across agents, skills, MCP servers, and rules to find optimization opportunities. Helps reduce token overhead and avoid performance warnings.

Analyze instincts and suggest or generate evolved structures

Export instincts from project/global scope to a file

Import instincts from file or URL into project/global scope

Show learned instincts (project + global) with confidence

Extract patterns from the current session + self-evaluate quality + auto-pick save location (project vs global scope). Slower but safer path with quality gating. For direct extraction use /learn.

Extract patterns from the current session as skills — direct save to the relevant location. Fast path. For quality gating + save-location decision use /learn-eval.

List known projects and their instinct statistics

Promote project-scoped instincts to global scope

Search cross-session memory for context, decisions, and prior work via the claude-mem MCP plugin.

Save an important fact to cross-session memory via the claude-mem MCP plugin.

Load the most recent session file from ~/.claude/session-data/ and resume work with full context from where the last session ended.

Save current session state to a dated file in ~/.claude/session-data/ so work can be resumed in a future session with full context.

Search and analyze session history — full-text search, frequency stats, pattern extraction across multiple sessions. For CRUD/management (list/load/alias/edit) use /sessions.

CRUD on Claude Code session files (list, load by date/alias, create alias, edit metadata) in `~/.claude/sessions/`. For search/stats/analytics across sessions use /session-analyzer.

Audits Claude Code context window consumption across agents, skills, MCP servers, and rules. Identifies bloat, redundant components, and produces prioritized token-savings recommendations.

Instinct-based learning system that observes sessions via hooks, creates atomic instincts with confidence scoring, and evolves them into skills/commands/agents. v2.1 adds project-scoped instincts to prevent cross-project contamination.

Operate and extend NanoClaw v2, ECC's zero-dependency session-aware REPL built on claude -p.

Scan skills to extract cross-cutting principles and distill them into rules — append, revise, or create new rule files

11 hooks across 5 events