add the validator script

Author: jrmolin

dev/docs-validator/.env.example

@@ -0,0 +1,3 @@
+MODEL = "gemini-2.5-flash"
+INTEGRATIONS_DIR = "~/git/integrations"
+#PACKAGE_NAME = "cisco_ise"

dev/docs-validator/.gitignore

@@ -0,0 +1,5 @@
+.env
+.env.*
+!.env.example
+
+.venv

dev/docs-validator/README.md

@@ -0,0 +1,31 @@
+# Docs Validator
+
+This script validates the `system_info.md` knowledge-base markdown file by:
+
+1.  Reading the markdown file and extracting URLs.
+2.  Validating the URLs.
+3.  Validating the markdown file's format against a template.
+
+## Installation
+
+This project uses `uv` for package management. To install the dependencies, run:
+
+```bash
+uv venv && uv sync
+```
+
+## Usage
+
+To run the script, execute the `validate.py` file:
+
+```bash
+uv run validate.py
+```
+
+## Environment Variables
+
+The script uses the following environment variables:
+
+*   `MODEL`: The model to use for the AI agents. Defaults to `google-gla:gemini-2.5-flash`.
+*   `INTEGRATIONS_REPO_PATH`: The path to the integrations repository. Defaults to the parent directory of the script.
+*   `PACKAGE_NAME`: The name of the package to validate. Defaults to `cisco_ftd`.

dev/docs-validator/_static/system_info_template.md

@@ -0,0 +1,53 @@
+
+# Service Info
+
+## Common use cases
+/* Common use cases that this will facilitate */
+
+## Data types collected
+/* What types of data this integration can collect */
+
+## Compatibility
+/* Information on the vendor versions this integration is compatible with or has been tested against */
+
+## Scaling and Performance
+/* Vendor-specific information on what performance can be expected, how to set up scaling, etc. */
+
+# Set Up Instructions
+
+## Vendor prerequisites
+/* Add any vendor specific prerequisites, e.g. "an API key with permission to access <X, Y, Z> is required" */
+
+## Elastic prerequisites
+/* If there are any Elastic specific prerequisites, add them here
+    The stack version and agentless support is not needed, as this can be taked from the manifest */
+
+## Vendor set up steps
+/* List the specific steps that are needed in the vendor system to send data to Elastic.
+  If multiple input types are supported, add instructions for each in a subsection */
+
+## Kibana set up steps
+/* List the specific steps that are needed in Kibana to add and configure the integration to begin ingesting data */
+
+# Validation Steps
+/* List the steps that are needed to validate the integration is working, after ingestion has started.
+    This may include steps on the vendor system to trigger data flow, and steps on how to check the data is correct in Kibana dashboards or alerts. */
+
+# Troubleshooting
+/* Add lists of "*Issue* / *Solutions*" for troubleshooting knowledge base into the most appropriate section below */
+
+## Common Configuration Issues
+/* For generic problems such as "service failed to start" or "no data collected" */
+
+## Ingestion Errors
+/* For problems that involve "error.message" being set on ingested data */
+
+## API Authentication Errors
+/* For API authentication failures, credential errors, and similar */
+
+## Vendor Resources
+/* If the vendor has a troubleshooting specific help page, add it here */
+
+# Documentation sites
+/* List of URLs that contain info on the service (reference pages, set up help, API docs, etc. */
+

dev/docs-validator/pyproject.toml

@@ -0,0 +1,15 @@
+[project]
+name = "validator-mcp"
+version = "0.1.0"
+description = "This project provides a tool for validating documentation markdown files. It checks for broken URLs and ensures that the files adhere to a predefined format template, leveraging AI to enhance the validation process."
+readme = "README.md"
+requires-python = ">=3.13"
+dependencies = [
+    "fastmcp>=2.13.0.2",
+    "langchain[google-genai]>=1.0.5",
+    "markdown-analysis>=0.1.5",
+    "markdown-it-pyrs>=0.4.0",
+    "pydantic-ai>=1.14.1",
+    "python-markdown>=0.1.0",
+    "rich>=14.2.0",
+]