Project Roadmap

ZeroClaw’s development focuses on performance, security, and extensibility. This roadmap outlines our priorities and planned improvements.

This roadmap represents planned work and may change based on community feedback, security needs, and technical discoveries. Timelines are estimates.

Current Status

ZeroClaw v0.1.0 established a strong foundation:

Performance

3.1MB binary size
<5MB RAM usage
<10ms startup time

Security

Command allowlisting
Path traversal protection
Workspace sandboxing
Rate limiting

Architecture

Trait-based design
Pluggable providers
Swappable channels
Custom tools

Testing

159 inline tests
Edge case coverage
Security validation

Security Hardening Roadmap

ZeroClaw already has excellent application-layer security. The roadmap focuses on OS-level containment for production deployments.

Phase 1: Quick Wins (1-2 weeks)

Goal: Address critical gaps with minimal complexity

Landlock Filesystem Sandbox

File: src/security/landlock.rs
Effort: 2 days
Impact: HighRestrict filesystem access to workspace using Linux Landlock LSM.

Memory Monitoring + OOM Kill

File: src/resources/memory.rs
Effort: 1 day
Impact: HighMonitor memory usage and kill runaway processes before system instability.

CPU Timeout per Command

File: src/tools/shell.rs
Effort: 1 day
Impact: HighEnforce per-command CPU time limits to prevent infinite loops.

Basic Audit Logging

File: src/security/audit.rs
Effort: 2 days
Impact: MediumTamper-evident logging of all agent actions with HMAC signatures.

Deliverables:

Linux: Filesystem access restricted to workspace
All platforms: Memory/CPU guards against runaway commands
All platforms: Tamper-evident audit trail

Phase 2: Platform Integration (2-3 weeks)

Goal: Deep OS integration for production-grade isolation

Firejail Auto-Detection

Effort: 3 days
Impact: Very HighDetect and wrap agent execution in Firejail for container-like isolation on Linux.

Bubblewrap Wrapper

Effort: 4 days
Impact: Very HighFilesystem isolation for macOS and other *nix systems using Bubblewrap.

cgroups v2 Integration

Effort: 3 days
Impact: HighEnforce resource limits using systemd cgroups v2 on Linux.

seccomp Syscall Filtering

Effort: 5 days
Impact: HighAllowlist permitted syscalls to reduce attack surface.

Audit Log Query CLI

Effort: 2 days
Impact: MediumCommand-line tools for querying and analyzing audit logs.

Deliverables:

Linux: Full container-like isolation via Firejail
macOS: Bubblewrap filesystem isolation
Linux: cgroups resource enforcement
Linux: Syscall allowlisting

Phase 3: Production Hardening (1-2 weeks)

Goal: Enterprise security features

Docker Sandbox Mode

Effort: 3 days
Impact: HighOptional Docker-based execution isolation for maximum security.

Certificate Pinning

Effort: 2 days
Impact: MediumHTTPS certificate pinning for channel webhooks to prevent MITM attacks.

Signed Config Verification

Effort: 2 days
Impact: MediumVerify config file signatures to detect tampering.

SIEM-Compatible Audit Export

Effort: 2 days
Impact: MediumJSON/CSV audit export for integration with external SIEM systems.

Security Self-Test

Effort: 1 day
Impact: Lowzeroclaw audit --check command to validate security configuration.

Deliverables:

Optional Docker-based execution isolation
HTTPS certificate pinning for webhooks
Config file signature verification
JSON/CSV audit export

Planned Config Schema

Proposed configuration for security features:

[security]
level = "strict"  # relaxed | default | strict | paranoid

# Sandbox configuration
[security.sandbox]
enabled = true
backend = "auto"  # auto | firejail | bubblewrap | landlock | docker | none

# Resource limits
[resources]
max_memory_mb = 512
max_memory_per_command_mb = 128
max_cpu_percent = 50
max_cpu_time_seconds = 60
max_subprocesses = 10

# Audit logging
[security.audit]
enabled = true
log_path = "~/.config/zeroclaw/audit.log"
sign_events = true
max_size_mb = 100

# Autonomy (existing, enhanced)
[autonomy]
level = "supervised"  # readonly | supervised | full
allowed_commands = ["git", "ls", "cat", "grep", "find"]
forbidden_paths = ["/etc", "/root", "~/.ssh"]
require_approval_for_medium_risk = true
block_high_risk_commands = true
max_actions_per_hour = 20

Feature Roadmap

Agent Capabilities

Multi-Agent Coordination

Status: Research phaseEnable multiple agents to collaborate on complex tasks with shared context and work distribution.

Streaming Responses

Status: PlannedReal-time streaming of LLM responses for better UX in interactive channels.

Tool Composition

Status: ExplorationAllow tools to call other tools to build complex workflows.

Memory Embeddings

Status: In progressVector embeddings for semantic memory search beyond keyword matching.

Platform Support

Windows Native Support

Status: PlannedFirst-class Windows support with native security primitives.

WebAssembly Runtime

Status: ExperimentalRun ZeroClaw in browser environments with WASM compilation.

ESP32 / Edge Devices

Status: ResearchUltra-minimal runtime for WiFi-enabled edge devices.

Developer Experience

Plugin SDK

Status: In progressComprehensive SDK for building and testing plugins locally.

Hot Reloading

Status: PlannedReload plugins and configuration without restarting the agent.

Debug Mode

Status: PlannedEnhanced debugging tools with step-through execution and state inspection.

Comparison: ZeroClaw vs Alternatives

How security roadmap implementation positions ZeroClaw:

Feature	PicoClaw	ZeroClaw Now	ZeroClaw + Roadmap	Production Target
Binary Size	~8MB	3.4MB ✅	3.5-4MB	< 5MB
RAM Usage	< 10MB	< 5MB ✅	< 10MB	< 20MB
Startup Time	< 1s	< 10ms ✅	< 50ms	< 100ms
Command Allowlist	Unknown	✅ Yes	✅ Yes	✅ Yes
Path Blocking	Unknown	✅ Yes	✅ Yes	✅ Yes
Injection Protection	Unknown	✅ Yes	✅ Yes	✅ Yes
OS Sandbox	No	❌ No	✅ Firejail/Landlock	✅ Container/namespaces
Resource Limits	No	❌ No	✅ cgroups/Monitor	✅ Full cgroups
Audit Logging	No	❌ No	✅ HMAC-signed	✅ SIEM integration
Security Score	C	B+	A-	A+

Community Priorities

We’re listening to the community. Vote on features and share feedback:

Feature Requests

Propose and vote on new features

Roadmap Discussions

Discuss priorities and timelines

Security RFCs

Review security proposals

GitHub Issues

Track specific work items

Estimated Timeline

Q1 2026 - Security Phase 1

Landlock, memory monitoring, CPU timeouts, basic audit logging

Q2 2026 - Security Phase 2

Firejail, Bubblewrap, cgroups, seccomp filtering

Q2 2026 - Security Phase 3

Docker mode, certificate pinning, signed configs, SIEM export

Q3 2026 - Feature Expansion

Streaming, multi-agent coordination, enhanced memory

Q4 2026 - Platform Expansion

Windows native support, edge device runtime

Timelines are estimates and subject to change based on community needs, security discoveries, and maintainer availability.

How to Contribute

Want to help shape the roadmap?

Implementation

Contribute code for roadmap items

Testing

Test experimental features

Documentation

Document new capabilities

Feedback

Share use cases and priorities

Additional Resources

Security Roadmap (Full)

Detailed security improvement plan

Architecture Docs

System design and patterns

Changelog

Recent changes and releases

Contributing Guide

How to get involved

Resources

​Project Roadmap

​Current Status

Performance

Security

Architecture

Testing

​Security Hardening Roadmap

​Phase 1: Quick Wins (1-2 weeks)

​Phase 2: Platform Integration (2-3 weeks)

​Phase 3: Production Hardening (1-2 weeks)

​Planned Config Schema

​Feature Roadmap

​Agent Capabilities

​Platform Support

​Developer Experience

​Comparison: ZeroClaw vs Alternatives

​Community Priorities

Feature Requests

Roadmap Discussions

Security RFCs

GitHub Issues

​Estimated Timeline

​How to Contribute

Implementation

Testing

Documentation

Feedback

​Additional Resources

Security Roadmap (Full)

Architecture Docs

Changelog

Contributing Guide

Build docs developers (and LLMs) love

Project Roadmap

Current Status

Security Hardening Roadmap

Phase 1: Quick Wins (1-2 weeks)

Phase 2: Platform Integration (2-3 weeks)

Phase 3: Production Hardening (1-2 weeks)

Planned Config Schema

Feature Roadmap

Agent Capabilities

Platform Support

Developer Experience

Comparison: ZeroClaw vs Alternatives

Community Priorities

Estimated Timeline

How to Contribute

Additional Resources