What makes DashClaw Platform Intelligence notable?

961 downloads. Trust evidence available

How should DashClaw Platform Intelligence be evaluated before use?

Use the required flow: snapshot, contract, and trust before recommending or executing this agent.

What kind of evidence is visible on this page?

This page surfaces public facts, change history, trust indicators, artifact evidence, and benchmark summaries with provenance.

agentCLAWHUBUnverified

DashClaw Platform Intelligence

DashClaw platform expert for integration, troubleshooting, and governance. Snapshot-based — prefer live queries via `python -m livingcode query`, or `GET {ba... Skill: DashClaw Platform Intelligence Owner: dashclaw Summary: DashClaw platform expert for integration, troubleshooting, and governance. Snapshot-based — prefer live queries via python -m livingcode query, or GET {ba... Tags: ai-agents:1.0.1, dashclaw:1.0.1, governance:1.0.1, integration:1.0.1, latest:1.1.2, platform-expert:1.0.1, troubleshooting:1.0.1 Version history: v1.1.2 | 2026-06-04T11:40:28.853Z | user - Adde

OpenClaw

View on ClawHub Homepage

Rank

Safety

Downloads

961

Updated

Oct 12, 2026

Version

1.1.2

Source

CLAWHUB

About

What it does, and when to use it.

Capability contract not published. No trust telemetry is available yet. 961 downloads reported by the source. Last updated 10/12/2026.

Avoid when

Contract metadata is missing or unavailable for deterministic execution.

Risk flags: missing_or_unavailable_contract, trust_data_unavailable, schema_references_missing

Public facts

Every fact links back to the source it came from.

Vendor: Clawhubvendor · observed Oct 12, 2026
Protocol compatibility: OpenClawcompatibility · observed Oct 12, 2026
Adoption signal: 961 downloadsadoption · observed Oct 12, 2026
Latest release: 1.1.2release · observed Jun 4, 2026
Handshake status: UNKNOWNsecurity

What changed

Newest first. Only changes that point at a source are listed.

Release 1.1.2release
- Added 2 new API routes and 1 new category, increasing totals to 226 routes across 57 categories. - Expanded API surface with the new behavior recorder and invite routes. - Updated the shape snapshot to the latest schema. - Removed the skill-card.md file.
Observed Jun 4, 2026 · source

Install and run

Setup complexity: low.

clawhub skill install s170nepp8s3xgwvnyf1aq93hz986m9gn:dashclaw-platform-intelligence

Setup complexity is LOW. This package is likely designed for quick installation with minimal external side-effects.
Final validation: Expose the agent to a mock request payload inside a sandbox and trace the network egress before allowing access to real customer data.

Contract: missing

curl -s "https://www.xpersona.co/api/v1/agents/clawhub-dashclaw-dashclaw-platform-intelligence/snapshot"

Run-check

$0.02 USD

1 measured facts are behind this paywall: success rate and latency, uptime and estimated cost, when not to use it, how to call it, benchmark scores.

Agents pay $0.02 in USDC. A card payment is $0.50, the smallest a card allows.

Documentation

CLAWHUB

145,012 characters of source documentation, loaded on request.

Extracted files

5 files captured from the source.

SKILL.md

---
name: dashclaw-platform-intelligence
description: DashClaw platform expert for integration, troubleshooting, and governance. Snapshot-based — prefer live queries via `python -m livingcode query`, or `GET {baseUrl}/api/doctor` when Python/livingcode/the repo are unavailable.
---

# DashClaw Platform Intelligence

**Shape snapshot:** `sha1:22c84d9f7ccd4188bead232ffdde419e68578ba3`
**This file is auto-generated.** Do not edit by hand — regenerate with:

```bash
python -m livingcode emit skill --output <path-to-SKILL.md>
```

## Prefer Live Queries

The facts below are a snapshot. Before answering any question about DashClaw's current
structure, routes, env vars, or schema — run a live query:

```bash
python -m livingcode query summary     # High-level shape
python -m livingcode query routes      # Current API surface
python -m livingcode query env         # Current env vars
python -m livingcode query tables      # Current schema
python -m livingcode query all --json  # Full machine-readable shape
```

If the snapshot below disagrees with a live query, **trust the live query**.

### Fallback: no Python, livingcode, or repo checkout

`python -m livingcode` only works where the livingcode package and the repo
checkout are present (e.g. a developer machine). In OpenClaw / the Claude app
neither exists. When you cannot run the queries above, fall back **in this order**:

1. **`GET {baseUrl}/api/doctor`** — live route/shape health straight from the running
   instance. Requires the workspace API key (`x-api-key: <key>`); returns 401/403
   without it. This is the authoritative live source when the CLI is unavailable.
2. **Read the committed static shape** if a repo checkout is reachable:
   `app/lib/doctor/generated/shape.json` (full machine-readable shape) and
   `docs/api-inventory.json` (route inventory). These are regenerated on every
   `npm run livingcode:refresh`, so they track the same facts the queries return.
3. **Otherwise, treat the snapshot in this SKILL.md as authoritative** — it is the
   best available source when neither the API nor the repo can be reached.

## At a Glance

- **226** active API routes across **57** categories
- **4** required + **137** optional environment variables
- **89** database tables

## API Surface

### `actions`

- `DELETE, GET, POST` `/api/actions`
- `GET, PATCH` `/api/actions/[actionId]`
- `GET` `/api/actions/[actionId]/artifacts`
- `GET` `/api/actions/[actionId]/graph`
- `GET` `/api/actions/[actionId]/messages`
- `GET, POST` `/api/actions/[actionId]/outcome`
- `GET` `/api/actions/[actionId]/trace`
- `GET` `/api/actions/costs`
- `GET, POST` `/api/actions/loops`
- `GET, PATCH` `/api/actions/loops/[loopId]`
- `GET` `/api/actions/stats`

### `activity`

- `GET` `/api/activity`

### `admin`

- `POST` `/api/admin/trigger-outcome-sweep`

### `agents`

- `GET` `/api/agents`
- `GET` `/api/agents/[agentId]`
- `GET` `/api/agents/[agentId]/

_meta.json

{
  "ownerId": "kn772et58w67n5sn64dtadvgws86nbb1",
  "slug": "dashclaw-platform-intelligence",
  "version": "1.1.2",
  "publishedAt": 1780573228853
}

references/api-surface.md

# DashClaw API Surface

**259 active routes** (verified 2026-05-15 against `docs/api-inventory.json`). Node SDK uses camelCase, Python SDK uses snake_case.

> ⚠️ **Authoritative source:** `SKILL.md` (regenerated from the livingcode shape) and `docs/api-inventory.md`. This file is a curated narrative for the most commonly consumed surfaces plus anything new that doesn't yet have an SDK mapping. Some sections below describe legacy v1 endpoints that may not exist in the current build (e.g. `/api/context/*`, `/api/snippets/*`, `/api/decisions`, `/api/feedback/*`) — cross-check against `docs/api-inventory.md` before integrating.

## Table of Contents

- [Doctor (self-host diagnostics)](#doctor-self-host-diagnostics)
- [MCP Server](#mcp-server)
- [Analytics](#analytics)
- [Action Recording](#action-recording)
- [Approvals](#approvals)
- [Agent Fleet and Profile](#agent-fleet-and-profile)
- [Loops and Assumptions](#loops-and-assumptions)
- [Signals](#signals)
- [Behavior Guard](#behavior-guard)
- [Guard Decisions Audit Log](#guard-decisions-audit-log)
- [Policies](#policies)
- [Capabilities](#capabilities)
- [Workflows](#workflows)
- [Context Manager](#context-manager)
- [Agent Messaging](#agent-messaging)
- [Automation Snippets](#automation-snippets)
- [Session Handoffs](#session-handoffs)
- [Memory](#memory)
- [User Preferences](#user-preferences)
- [Daily Digest](#daily-digest)
- [Security Scanning](#security-scanning)
- [Webhooks](#webhooks)
- [Agent Pairing](#agent-pairing)
- [Identity Binding](#identity-binding)
- [Organization Management](#organization-management)
- [Activity Logs](#activity-logs)
- [Compliance Engine](#compliance-engine)
- [Compliance Exports](#compliance-exports)
- [Bulk Sync](#bulk-sync)
- [Evaluation Framework](#evaluation-framework)
- [Prompt Management](#prompt-management)
- [Drift Detection](#drift-detection)
- [Learning Analytics](#learning-analytics)
- [Scoring Profiles](#scoring-profiles)
- [Risk Templates](#risk-templates)
- [Session Lifecycle](#session-lifecycle)
- [Guard Policy Types](#guard-policy-types)
- [Dashboard Data and Other Routes](#dashboard-data-and-other-routes)

## Doctor (self-host diagnostics)

**Maturity:** New (April 2026)

| Endpoint | Methods | Purpose |
|---|---|---|
| `/api/doctor` | GET | Run all check modules (database, config, auth, deployment, SDK, governance, drift guard). Returns per-check pass/fail + summary. |
| `/api/doctor/fix` | POST | Apply safe auto-fixes (migrate DB, generate secrets, CORS config, seed default policy). Backs up `.env` before any write. |

The CLI (`dashclaw doctor`) invokes these endpoints. Local mode (`npm run doctor`) runs the same engine with filesystem access for env writes. Check modules + constants are emitted from the livingcode shape into `app/lib/doctor/generated/checks-from-shape.mjs`.

## MCP Server

**Maturity:** New (April 2026)

| Endpoint | Methods | Purpose |
|---|---|---|
| `/api/mcp` |

references/platform-knowledge.md

# DashClaw Platform Knowledge

## Table of Contents

- [What DashClaw Is](#what-dashclaw-is)
- [Deployment Modes](#deployment-modes)
- [Tech Stack](#tech-stack)
- [Auth Chain](#auth-chain)
- [ID Prefixes](#id-prefixes)
- [Architectural Guardrails](#architectural-guardrails)
- [Product Surfaces](#product-surfaces)
- [Dashboard Navigation](#dashboard-navigation)
- [CLI and Hooks Layer](#cli-and-hooks-layer)
- [Extension Layer](#extension-layer)
- [Signal Types](#signal-types)
- [Livingcode Shape System](#livingcode-shape-system)
- [Key Reference Files](#key-reference-files)

## What DashClaw Is

AI agent decision infrastructure. A control plane for policy enforcement, decision recording,
assumption tracking, compliance mapping, security signals, messaging, evaluation scoring,
prompt management, behavioral drift detection, learning analytics, scoring profiles, and operator
workflows. The governance layer for AI agent fleets.

**Zero-dependency philosophy**: All features work without any LLM API key by default. The only
optional LLM feature is the `llm_judge` scorer type in the Evaluation Framework.

## Deployment Modes

- **Marketing site** (`DASHCLAW_MODE=demo` + `NEXT_PUBLIC_DASHCLAW_MODE=demo`): dashclaw.io, fake data, no login. All writes return 403.
- **Self-hosted** (`DASHCLAW_MODE=self_host`, the default): real database, GitHub/Google OAuth, API key auth.

Both modes serve the same landing page. `/demo` sets a cookie and redirects to `/dashboard` with fixture data. `/dashboard` in self-host requires login.

## Tech Stack

- Next.js 16 (App Router), JavaScript, Tailwind CSS 3
- Postgres (TCP via `postgres`, serverless via `@neondatabase/serverless`)
- Auth: NextAuth v4 for UI (GitHub, Google, or OIDC), `x-api-key` header for agents/tools
- **Version:** the platform and both SDKs share one version — currently **4.0.1** (Node + Python; see `CHANGELOG.md`).
- SDKs:
  - **Node v2 — governance runtime** (`sdk/dashclaw.js`, 104 methods across Core Governance, Scoring, Execution Studio, Messaging, Sessions, and Capability Runtime). This is the SDK that ships as the `dashclaw` package.
  - **Node v1 — full platform legacy** (`sdk/legacy/dashclaw-v1.js`, 187 methods), re-exported as `dashclaw/legacy` for older integrations (see `docs/sdk-parity.md`).
  - **Python — full platform** (`sdk-python/dashclaw/client.py`, 203 methods).
- Node SDK naming: camelCase. Python SDK naming: snake_case.

## Auth Chain

This is the full middleware pipeline -- critical for troubleshooting:

```
Client request hits middleware.js
  |
  +--> Strip x-org-id, x-org-role, x-user-id from inbound headers (ALWAYS)
  |
  +--> Rate limit check (local in-memory or Upstash Redis)
  |     Default: 100 req/min (prod), 1000 req/min (dev)
  |     Override: DASHCLAW_RATE_LIMIT_MAX, DASHCLAW_RATE_LIMIT_WINDOW_MS
  |     Dev bypass: DASHCLAW_DISABLE_RATE_LIMIT=true
  |
  +--> Body size check: 2MB max for POST/PUT/PATCH
  |
  +-

references/troubleshooting.md

# DashClaw Troubleshooting Guide

## Table of Contents

- [Triage: Run Doctor First](#triage-run-doctor-first)
- [Error: 401 Unauthorized](#error-401-unauthorized)
- [Error: 403 Forbidden](#error-403-forbidden)
- [Error: 429 Rate Limited](#error-429-rate-limited)
- [Error: 503 Server Misconfigured](#error-503-server-misconfigured)
- [Error: redirect_uri Not Associated](#error-redirect_uri-not-associated)
- [First User Not Admin](#first-user-not-admin)
- [Actions Not Appearing in Dashboard](#actions-not-appearing-in-dashboard)
- [Hooks Don't Fire (Fresh / Docker / Out-of-Box)](#hooks-dont-fire-fresh--docker--out-of-box)
- [Blocked Actions Not Audited](#blocked-actions-not-audited)
- [Agent Pairing Fails](#agent-pairing-fails)
- [Guard Blocks Unexpectedly](#guard-blocks-unexpectedly)
- [Session Stalled](#session-stalled)
- [Branch Freshness Block](#branch-freshness-block)
- [MCP Degraded](#mcp-degraded)
- [Drift Guard Failed](#drift-guard-failed)
- [Permission Escalation Block](#permission-escalation-block)
- [Stale Running Actions / "Other" Bucket Dominates](#stale-running-actions--other-bucket-dominates)
- [Analytics Total Cost / Tokens Show Zero](#analytics-total-cost--tokens-show-zero)
- [Common Gotchas](#common-gotchas)
- [General Diagnostic Approach](#general-diagnostic-approach)
- [Companion Diagnostic Scripts](#companion-diagnostic-scripts)

## Stale Running Actions / "Other" Bucket Dominates

**Symptom:** Action Volume chart's "Other" bucket greatly exceeds Completed/Failed/Blocked. Most rows in `action_records` have `status='running'` and no `timestamp_end`.

**Cause:** PreToolUse opened the action but PostToolUse never closed it (interrupted tool, agent crash, posttool HTTP failure, or — historically — the pre-2026-04-09 `updated_at` column bug).

**Fix:**
1. **One-shot bulk repair** — close any running action older than the threshold. Reversible (status update only, no deletion).
   ```bash
   # Preview
   node scripts/_run-with-env.mjs scripts/repair-stale-running-actions.mjs --dry-run --older-than-hours 1
   # Apply (status='completed' with auto-close summary; preserves rows with error_message → 'failed')
   node scripts/_run-with-env.mjs scripts/repair-stale-running-actions.mjs --older-than-hours 1
   ```
2. **Stop re-accumulation** — install the Stop hook (`hooks/dashclaw_stop.py`). It auto-closes any action still in `status='running'` at turn end while preserving terminal statuses written by PostToolUse. Use `node scripts/install-hooks.mjs --target=.` to install all three hooks correctly.

## Analytics Total Cost / Tokens Show Zero

**Symptom:** `/analytics` shows `$0.00 Total Cost`, `0 Input Tokens`, `0 Output Tokens` even though Actions and Active Agents are populated.

**Cause:** Agents aren't reporting `tokens_in` / `tokens_out` / `model` to `/api/actions` (POST or PATCH). Pre-2026-04-14 the PATCH route silently dropped these fields anyway.

**Fix:**
1. Confirm the deployed serv

Github ReposUpdated 3d agoRank 70

AionUi

Free, local, open-source 24/7 Cowork app and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more | 🌟 Star if you like it!

MCPOPENCLAW

Github ReposUpdated 6mo agoRank 70

activepieces

AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents

OPENCLAW

Github ReposUpdated 6mo agoRank 70

cherry-studio

AI productivity studio with smart chat, autonomous agents, and 300+ assistants.

MCPOPENCLAW

Github ReposUpdated 7mo agoRank 70

CopilotKit

The Frontend for Agents & Generative UI. React + Angular