← Back to Skills Marketplace
Haresh Product Search
by
Haresh Sainaath S
· GitHub ↗
· v1.0.2
425
Downloads
0
Stars
1
Active Installs
3
Versions
Install in OpenClaw
/install haresh-product-search
Description
Search e-commerce products via n8n webhook integration
README (SKILL.md)
Product Search Skill
Purpose
Enables users to search for products in the e-commerce catalog using natural language queries.
When to Use
- User asks to find products, search inventory, or browse catalog
- User mentions specific product types, categories, or brands
- User provides price constraints or filtering requirements
Workflow
-
Parse user intent to extract search parameters:
- Product category (e.g., "running shoes", "laptops")
- Price constraints (e.g., "under $100", "between $50-$200")
- Sort preferences (e.g., "cheapest first", "highest rated")
-
Transform parameters into JSON payload with category, price_min, price_max, sort_by
-
Use exec tool to POST to n8n webhook at http://localhost:5678/webhook/product-search
-
Parse n8n response and present results to user in friendly format
Parameter Mapping
- Category synonyms: "gym shoes" becomes "fitness", "sneakers" becomes "footwear"
- Price parsing: "under X" sets price_max to X
- Sort options: "cheapest" sorts price ascending, "best" sorts by rating
Error Handling
- If no results: Suggest broader search terms
- If webhook fails: Inform user search service is unavailable
- If timeout: Ask if user wants to wait or retry\r
Usage Guidance
This skill is internally inconsistent and should not be used until corrected. Key things to check or require from the author before installing:
- Provide a configurable webhook URL (e.g., N8N_WEBHOOK_URL) declared in requires.env instead of a hard-coded placeholder.
- Resolve the mismatch between SKILL.md (which instructs an exec POST to localhost) and index.js (which does a fetch to an external placeholder). Confirm whether the webhook target is meant to be localhost or an external n8n instance.
- If the webhook will be an external endpoint, verify the endpoint is trusted and requires authentication; do not allow the skill to post arbitrary user data to an untrusted third party.
- Prefer returning the tool behavior in code and keep SKILL.md instructions aligned and specific (avoid open-ended exec guidance that could run arbitrary shell commands).
- Ask the author for a new release with proper configuration, documented auth requirements (if any), and matching docs/code before enabling the skill. If you must test, run it in an isolated environment with network egress controls so you can observe and restrict where it posts.
Capability Analysis
Type: OpenClaw Skill
Name: haresh-product-search
Version: 1.0.2
The SKILL.md file instructs the AI agent to use an 'exec' tool to make a POST request to 'http://localhost:5678/webhook/product-search'. This explicit instruction to interact with a local address via an 'exec' tool represents a potential Server-Side Request Forgery (SSRF) or local service interaction vulnerability. While the intent might be to connect to a legitimate local n8n instance, this capability, if the agent's execution environment is not properly sandboxed, could be exploited to access or manipulate other services on the host. The index.js file contains a placeholder URL, which is not directly malicious but highlights the skill's incomplete configuration.
Capability Assessment
Purpose & Capability
The skill claims to integrate with an n8n webhook (reasonable for a product-search skill). However, there is no declared or required webhook URL or configuration. The code hard-codes a placeholder URL (https://your-n8n-webhook-url) instead of reading a configurable environment variable, so the skill as packaged cannot reliably reach the intended service. That omission is disproportionate to the stated purpose and suggests incomplete/incorrect packaging.
Instruction Scope
SKILL.md tells the agent to use an exec tool to POST to http://localhost:5678/webhook/product-search, while the actual runtime code (index.js) performs a fetch to a different placeholder URL (https://your-n8n-webhook-url). This conflict is material: following SKILL.md would try to contact localhost, while invoking the provided tool runs code that contacts an external URL. The instructions do not request reading unrelated files or secrets, but the mismatch creates ambiguity about outbound network targets (local vs external), which is a security-relevant inconsistency.
Install Mechanism
No install spec is provided (instruction-only with a small code file). That minimizes disk-install risk — nothing is downloaded from remote installers during installation.
Credentials
The skill declares no required environment variables or credentials, but its code needs a webhook URL to function and instead contains a hard-coded placeholder. A properly designed webhook integration would require a configurable URL (and possibly an API key) declared in requires.env. The lack of declared configuration means either the package is incomplete or it expects a developer to edit the code, both of which are problematic from a security/operational standpoint.
Persistence & Privilege
The skill does not request persistent/all-skill privileges (always:false) and does not modify other skills. Normal autonomous invocation is allowed by default but is not combined with other red flags here.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install haresh-product-search - After installation, invoke the skill by name or use
/haresh-product-search - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.2
- Initial implementation of haresh-product-search skill.
- Enables natural language product searches via n8n webhook integration.
- Added support for parsing product categories, price constraints, and sorting preferences.
- Returns user-friendly results or appropriate error messages if search fails or yields no results.
v1.0.1
- Added a _meta.json metadata file for improved skill management and compatibility.
- No changes to existing skill functionality or documentation.
v1.0.0
Initial release of haresh-product-search skill
- Enables users to search e-commerce products using natural language queries.
- Supports product category, price constraints, and sorting preferences.
- Integrates with an n8n webhook for backend search processing.
- Handles errors gracefully with user-friendly messaging.
Metadata
Frequently Asked Questions
What is Haresh Product Search?
Search e-commerce products via n8n webhook integration. It is an AI Agent Skill for Claude Code / OpenClaw, with 425 downloads so far.
How do I install Haresh Product Search?
Run "/install haresh-product-search" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Haresh Product Search free?
Yes, Haresh Product Search is completely free (open-source). You can download, install and use it at no cost.
Which platforms does Haresh Product Search support?
Haresh Product Search is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Haresh Product Search?
It is built and maintained by Haresh Sainaath S (@haresh-sai06); the current version is v1.0.2.
More Skills