← Back to Skills Marketplace
xiaojiou176

DealWatch Read-only Builder

by Yifeng[Terry] Yu · GitHub ↗ · v1.1.1 · MIT-0
cross-platform ✓ Security Clean
157
Downloads
0
Stars
0
Active Installs
3
Versions
Install in OpenClaw
/install dealwatch-readonly-builder
Description
Teach the agent how to connect the published DealWatch MCP package, choose the right read-only tool, and guide the user through the compare-first safe path w...
README (SKILL.md)

DealWatch Read-only Builder

Use this skill when the user wants the shortest truthful path into DealWatch.

What this skill teaches

  • how to install the published DealWatch MCP package
  • how to choose the right read-only tool for the current job
  • how to start with runtime readiness and compare preview before deeper reads
  • how to keep the answer inside the current local-first boundary

When to use this skill

Use this skill when the user asks to:

  • connect DealWatch to OpenHands or OpenClaw
  • check whether DealWatch is ready for a local read-only session
  • compare grocery product URLs before creating durable state
  • inspect watch tasks, watch groups, or recovery state without mutating anything

If the MCP is not connected yet

Open README.md in this folder and follow references/INSTALL.md. Do not invent repo-local launch commands when the published package already exists.

Safe-first workflow

  1. get_runtime_readiness Use this first when setup, environment, or startup truth is unclear.
  2. compare_preview Use this first when the user wants to compare candidate grocery URLs without creating tasks.
  3. get_builder_starter_pack Use this when the user needs the integration contract or host setup path.
  4. list_watch_tasks or list_watch_groups Use these after the first-safe path is already clear.
  5. get_watch_task_detail or get_watch_group_detail Use these for one deeper read-only follow-up.
  6. get_recovery_inbox, get_notification_settings, or get_store_onboarding_cockpit Use these only after the user is already inside a recovery or operator diagnostic workflow.

Tool-selection rule

  • Choose compare_preview for “Which URL is the right target?”
  • Choose get_runtime_readiness for “Is the local runtime healthy?”
  • Choose get_builder_starter_pack for “How do I connect an agent or host?”
  • Choose watch-task/group reads only after the user already has durable state to inspect

What to return

Return a short answer with:

  1. the chosen lane
  2. the next 1-3 actions
  3. one boundary reminder
  4. one exact MCP tool or install snippet

Guardrails

  • write-side MCP
  • hosted control plane
  • official marketplace or registry listing
  • autonomous recommendation support

Read next

  • references/INSTALL.md
  • references/CAPABILITIES.md
  • references/DEMO.md
  • references/TROUBLESHOOTING.md
Usage Guidance
This package appears to be a safe, coherent instruction-only skill that guides use of the published dealwatch==1.0.1 MCP in read-only mode. Before installing, verify the referenced PyPI package (dealwatch==1.0.1) is the correct, trusted release and that the uvx tool is available on your host. If you need stronger assurance, inspect the upstream dealwatch package source (or its PyPI/GitHub release) to confirm it behaves as documented. Note the skill does not request credentials or claim write/hosted control-plane capabilities; keep following the read-only boundary when using it.
Capability Analysis
Type: OpenClaw Skill Name: dealwatch-readonly-builder Version: 1.1.1 The DealWatch Read-only Builder skill bundle provides instructions and configuration for an agent to interact with a grocery price comparison MCP server. It emphasizes a safe-first, read-only workflow, using standard 'uvx' commands to execute a published PyPI package (dealwatch==1.0.1). The bundle contains no executable code, obfuscation, or malicious instructions, and explicitly defines boundaries to prevent unauthorized write operations or data access.
Capability Assessment
✓ Purpose & Capability
Name/description match the content: all files and SKILL.md consistently describe guiding an agent to use the published dealwatch==1.0.1 MCP in read-only mode. Required resources (none) and instructions (use published package via uvx) are proportional to that purpose.
✓ Instruction Scope
Runtime instructions stay within the defined read-only and local-first boundary. They prescribe smoke checks and which MCP tools to call; they do not instruct reading unrelated host files, harvesting env vars, or posting data to external endpoints.
✓ Install Mechanism
This is instruction-only and points to a published PyPI package (dealwatch==1.0.1) invoked via the uvx wrapper. No archive downloads, third-party URLs, or extract steps are present in the bundle.
✓ Credentials
The skill requests no environment variables, no credentials, and no config paths beyond providing example MCP config snippets for host wiring. That aligns with a read-only connector guide.
✓ Persistence & Privilege
The manifest does not set always:true and does not request persistent privileges or modify other skills. Autonomous invocation is allowed by default (platform normal) but is not combined here with broad access.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install dealwatch-readonly-builder
  3. After installation, invoke the skill by name or use /dealwatch-readonly-builder
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.1.1
- Added full reference documentation under the new references/ directory: INSTALL.md, CAPABILITIES.md, DEMO.md, TROUBLESHOOTING.md, and sample configuration files. - Updated SKILL.md to reference new documentation paths; installation instructions now point to INSTALL.md. - Expanded "Read next" section in SKILL.md to include new reference files. - Updated README.md and manifest.yaml as part of the documentation reorganization.
v1.1.0
- Expanded guidance in SKILL.md on how and when to use the skill, including clearer tool-selection rules and workflows. - Added practical instructions for MCP installation and tool choices. - Introduced new triggers and updated usage scenarios in SKILL.md. - Added three reference files: example tasks, MCP install, and tool map. - Improved boundary/reminder language to prevent unsupported use cases.
v1.0.0
- Initial release of dealwatch-readonly-builder. - Provides a safe-first builder route for DealWatch with a local-first, read-only approach. - Includes basic flows: runtime readiness check, builder starter pack, preview comparison, and a single read-only detail route. - Designed with clear boundaries: no write-side MCP, control plane hosting, registry listing, or autonomous recommendations.
Metadata
Slug dealwatch-readonly-builder
Version 1.1.1
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 3
Frequently Asked Questions

What is DealWatch Read-only Builder?

Teach the agent how to connect the published DealWatch MCP package, choose the right read-only tool, and guide the user through the compare-first safe path w... It is an AI Agent Skill for Claude Code / OpenClaw, with 157 downloads so far.

How do I install DealWatch Read-only Builder?

Run "/install dealwatch-readonly-builder" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is DealWatch Read-only Builder free?

Yes, DealWatch Read-only Builder is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does DealWatch Read-only Builder support?

DealWatch Read-only Builder is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created DealWatch Read-only Builder?

It is built and maintained by Yifeng[Terry] Yu (@xiaojiou176); the current version is v1.1.1.

More Skills
self-improving agent
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Clau...
⬇ 463556 · by pskoett
Skill Vetter
Security-first skill vetting for AI agents. Use before installing any skill from ClawdHub, GitHub, or other sources. Checks for red flags, permission scope, and suspicious patterns.
⬇ 259610 · by spclaudehome
Polymarket
Query Polymarket prediction markets. Check odds, find trending markets, search events, track price movements.
⬇ 232503 · by joelchance
Self-Improving + Proactive Agent
Self-reflection + Self-criticism + Self-learning + Self-organizing memory. Agent evaluates its own work, catches mistakes, and improves permanently. Use when...
⬇ 200551 · by ivangdavila
Github
Interact with GitHub using the `gh` CLI. Use `gh issue`, `gh pr`, `gh run`, and `gh api` for issues, PRs, CI runs, and advanced queries.
⬇ 191226 · by steipete
ontology
Typed knowledge graph for structured agent memory and composable skills. Use when creating/querying entities (Person, Project, Task, Event, Document), linkin...
⬇ 190235 · by oswalpalash

💬 Comments