#4646th of 72 in Gateways and observability

dashclaw

Approval and policy layer that blocks risky agent tool calls before they run

Releases and problems feed (RSS)

Stars
320
License
MIT
Last commit
Nov 2026
Last release
Nov 2026
Language
TypeScript
Stars, last 29 weeks+73
30 Apr 2026: 247 stars7 May 2026: 252 stars14 May 2026: 261 stars21 May 2026: 262 stars28 May 2026: 266 stars4 Jun 2026: 271 stars11 Jun 2026: 272 stars18 Jun 2026: 274 stars25 Jun 2026: 275 stars2 Jul 2026: 279 stars9 Jul 2026: 282 stars16 Jul 2026: 282 stars23 Jul 2026: 284 stars30 Jul 2026: 285 stars6 Aug 2026: 290 stars13 Aug 2026: 294 stars20 Aug 2026: 297 stars27 Aug 2026: 298 stars3 Sept 2026: 298 stars10 Sept 2026: 304 stars17 Sept 2026: 307 stars24 Sept 2026: 310 stars1 Oct 2026: 311 stars8 Oct 2026: 312 stars15 Oct 2026: 314 stars22 Oct 2026: 315 stars29 Oct 2026: 315 stars5 Nov 2026: 317 stars12 Nov 2026: 319 stars19 Nov 2026: 320 stars

247 on 30 Apr 2026320 on 19 Nov 2026

From GitHub's daily star counts.

Checked on GitHub.

Overview

DashClaw scores each agent tool call against policies and returns allow, warn, allow_contained, require_approval or block. Hooks for Claude Code, Codex and Hermes, plus the OpenClaw gateway, can cancel the call, and approvals can be resolved remotely from a web inbox, CLI, PWA, Telegram or Discord. Decisions go to a ledger with Ed25519 receipts where issued. SDK, REST and generic MCP integrations are cooperative only.

Who it is for: Developers running unattended coding agents on real repos and infrastructure

Strengths

  • Fail-closed PreToolUse hooks (exit 2) for Claude Code, Codex and Hermes
  • Remote approval via web inbox, CLI, phone PWA, Telegram or Discord
  • Signed decision ledger with Ed25519 receipts and signed exports
  • `npx dashclaw up` provisions Postgres, secrets, API key and hooks locally

Weaknesses

  • Enforcement is cooperative for bare SDK, REST and ordinary MCP callers
  • Hook runs at the agent's privilege; tamper resistance needs container or separate OS user
  • First commit February 2026; README says it is not battle-tested at scale
  • Day-to-day maintenance is done by an AI agent under a human charter

What it needs

  • no GPU
  • Docker + Compose
  • Needs Postgres (Docker or embedded), Node 20+
  • port 3000

What changed

From the release notes. No flag means the note does not say so, not that the upgrade is safe.

Also in Gateways and observability

See all 72
Also in Gateways and observability
RankProjectScore
1OmniRouteLLM gateway routing one OpenAI-compatible endpoint across 372 providers89.3k starsMIT96 out of 100
2headroomCompresses tool outputs, logs and history before they reach the LLM80.2k stars+3,058 (+4.0%) in 30 daysApache-2.092 out of 100
3MLflowTracing, evaluation and prompt registry for LLM apps, plus ML tracking Live demo ↗ (opens in a new tab)28.9k stars+478 (+1.7%) in 30 daysApache-2.092 out of 100
#4new-apiGateway that unifies LLM providers behind OpenAI, Anthropic and Gemini APIs52.3k stars+1,968 (+3.9%) in 30 daysAGPL-3.0known advisory in latest version91 out of 100
#5LiteLLMOpenAI-format gateway and Python SDK for calling 100+ LLM providers63.6k starscustom licensepart of the code needs a paid license88 out of 100