
Declaw Ai
io.github.declaw-aiPublished by GitHub account declaw-aiv0.1.2Updated Oct 23, 2026
Secure Firecracker microVM sandboxes for AI agents: network policy, PII & injection guardrails.
Overview
Lets an assistant create and operate secure Firecracker microVM sandboxes with network policies, PII scanning, and injection guardrails.
- What you can do
- Provides tools to create_sandbox, run_command, read_file, write_file, list_files, kill_sandbox, and list_sandboxes (R10-R16). Sandboxes are created with a security preset: none, standard (PII scanning plus audit logging), or strict (PII scanning, prompt injection defense, audit logging, and network deny-all), and allowed_domains can restrict outbound traffic (R18-R24). Eight built-in templates are offered, along with snapshots and interactive stdio (R32-R34).
- When to use it
- Use it when an assistant needs to run untrusted or generated code, shell commands, or file operations in an isolated environment rather than on the host. It suits workflows that need network restrictions, PII scanning, injection defense, or an audit trail around agent execution (R2, R23).
- Requirements
- Runs locally over stdio via npx @declaw/mcp-server (R6, R7). Requires Node.js and network access to the Declaw API. A Declaw API key is required in DECLAW_API_KEY (V1.1, R37); DECLAW_DOMAIN is optional for self-hosted deployments (V2.1, R38). Desktop only; no web executable.
Installation
In SourceWeft
- Open Declaw Ai in the dashboard and add it to a workspace.
- Enable the server for the chats that should use its tools.
Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.
Other MCP clients
Follow the launch instructions in the repository.
README
This repository has no README.
Tools
0Version history
1- v0.1.2LatestSep 16, 2026
Score
25% of the applicable checks met
How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.
- Reached in the last 30 days25 pts
Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.
- Auth declared10 pts
Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.
- Tools listed15 pts
Answer tools/list without a credential, or ship a package that starts in a sandbox without one.
- DescriptionRequired15 pts
- Logo10 pts
- Documentation15 pts
Declare a GitHub repository with a README, or serve resource_documentation in the OAuth protected-resource metadata.
- License10 pts
Declare a license in the repository or the registry entry.