Skip to content

Features

Jason L. West edited this page Feb 6, 2026 · 3 revisions

Features

Nebulus Atom includes 30+ features organized into core capabilities, developer tools, advanced features, and Overlord meta-orchestration.

Core Capabilities

Context Manager

Pin files to the active conversation so the agent always has them in context. Supports adding, removing, and listing pinned files.

Smart Undo

Automatically creates checkpoints before risky file operations (writes, edits, command execution). Restore any checkpoint to undo changes.

Skill Library

Persistent, reusable capabilities that the agent can learn and execute. Skills are saved to disk and available across sessions.

Semantic Code Search (RAG)

Uses ChromaDB and sentence-transformers to build embeddings of your codebase. Search for code by meaning, not just text patterns.

Autonomous Execution

Run the agent in autonomous mode where it executes multiple turns without waiting for confirmation. Includes safety limits.

Agentic TDD

Test-driven development automation. Give the agent a goal and it cycles through: write test, run test (fail), write code, run test (pass), refactor.

Developer Tools

Codebase Cartographer

AST-based code analysis that builds a map of your codebase structure: classes, functions, imports, and their relationships.

Shell Macro Generator

Generate and save shell macros from natural language descriptions. Macros are stored for reuse.

Sandbox Execution

Run shell commands in a sandboxed environment that restricts file system access to safe directories.

MCP Client

Model Context Protocol integration for connecting to external tool servers and expanding the agent's capabilities.

Session Journal

Automatic logging of session activities, decisions, and outcomes for review and learning.

Advanced Features

Advanced Cognition

"System 2" thinking that triggers deeper reasoning for complex tasks. The agent recognizes when a problem needs more careful analysis.

Adaptive Preference Learning

The agent learns your preferences over time: coding style, naming conventions, tool preferences. Stored per-session.

Interactive Clarification

When requirements are ambiguous, the agent asks clarifying questions before proceeding rather than making assumptions.

Multimodal Input

Support for image and other non-text inputs when the underlying model supports it.

Task Management

Structured task tracking with plans, subtasks, and progress monitoring.

Visual Task Graph

Visualize task dependencies and progress in the TUI dashboard.

Telemetry and Monitoring

Logging

Structured logging with configurable levels and output formats (console, JSON, file).

Flight Recorder Dashboard

Streamlit-based dashboard showing tool usage, response times, token consumption, and session history.

GitHub Integration

Tools for interacting with GitHub: reading issues, creating PRs, managing labels.

Overlord Meta-Orchestration

Project Registry & Scanner (Phase 1)

Multi-repo discovery via overlord.yml with dependency tracking, topological ordering (Kahn's algorithm), and circular dependency detection. Health scanner monitors git state, test health, and branch status across all projects.

Dependency Graph (Phase 1)

DAG traversal for upstream/downstream analysis, blast radius calculation, affected project sets, and ASCII tree rendering. Used by dispatch and autonomy engines to understand change impact.

Cross-Project Memory (Phase 1)

SQLite-backed observation store for remembering patterns, gotchas, and decisions across projects. Supports search by keyword/project/category, confidence tagging, and automatic pruning.

Autonomy Engine (Phase 2)

Confidence-based auto-dispatch with cautious/proactive/scheduled levels. Scores changes on scope, test health, and observation history. Per-project overrides and pre-approved action lists.

Multi-Repo Dispatch (Phase 2)

Executes tasks across projects in dependency order with rollback support. TaskParser converts natural language commands into DispatchPlans with scope analysis.

Slack Command Router (Phase 3)

Multi-project Slack commands via @atom mentions: status, scan, merge, release, autonomy, memory, approve, deny, help. Bridges async Slack handlers to sync Phase 2 modules via asyncio.to_thread().

Proposal Manager (Phase 3)

Approval workflow with Slack thread binding. Proposals flow through: pendingapprovedexecutingcompleted (or denied/expired). Thread replies control execution. SQLite-backed persistent store with auto-expiry.

Background Daemon (Phase 3)

Persistent process with croniter-based scheduling. Runs hourly health scans, nightly test sweeps, and weekly stale branch detection. Graceful shutdown via SIGINT/SIGTERM. Automatic Slack bot connection.

Proactive Detection (Phase 3)

Three built-in detectors run during scheduled scans:

  • StaleBranchDetector — branches with no activity for N days
  • AheadOfMainDetector — develop branches with unmerged commits
  • FailingTestDetector — projects with test failures or missing tests

Results are filtered by autonomy level and routed to Slack or the proposal system.

Notification System (Phase 3)

Dual-mode alert routing: urgent notifications post immediately to Slack, while routine events accumulate into a daily digest. Categories tracked: detections, proposals created/approved/denied, executions, health checks, test sweeps.

Feature Documentation

Detailed specifications for each feature are available in the repository at docs/features/. Each feature document includes:

  • Overview and motivation
  • Requirements
  • Technical implementation details
  • Verification plan

Related Pages

Clone this wiki locally