Docs for Agents
Ping Identity documentation supports both human readers and AI agents.
Every documentation page has a machine-readable Markdown alternate, and every domain and product docset exposes an llms.txt index so agents can efficiently discover and load the right content.
Documentation domains
Ping Identity documentation is served from two domains:
| Domain | What it covers |
|---|---|
Developer-focused content: Model Context Protocol (MCP) servers, agent skills, SDKs, and AI-first identity tooling. |
|
Product documentation for all Ping Identity products: PingAM, PingFederate, PingDirectory, and more. |
Markdown alternates
Every documentation page has a Markdown alternate, which is a clean, token-efficient version of the same content, stripped of HTML, navigation chrome, and other rendering artifacts. Markdown alternates reduce token usage compared to raw HTML and improve accuracy when used as agent context.
How to access a Markdown alternate
You can retrieve the Markdown alternate for any page using one of the following methods:
| Method | How to use it |
|---|---|
URL extension swap |
Replace |
HTML |
Every page includes a |
llms.txt |
Each product docset’s |
View Markdown link |
On any documentation page, a View Markdown link opens the Markdown alternate directly. This is primarily for human use. The previous methods are more suitable for programmatic access. |
llms.txt endpoints
Ping Identity documentation exposes llms.txt files following the llms.txt standard.
These files provide a structured index of documentation that agents and large language models (LLMs) can use to discover relevant content without crawling the full site.
Site-wide endpoints
| URL | What it contains |
|---|---|
Index of all developer-facing content on developer.pingidentity.com, organized by product area. |
|
Index of all product documentation on docs.pingidentity.com, organized by product docset. |
Per-docset endpoints
Each product docset also exposes its own scoped llms.txt, so an agent focused on a specific product can load only what it needs.
| URL pattern | Example |
|---|---|
|
|
|
Replace {product} with the docset identifier for the product you are targeting.
Valid product identifiers are listed in the site-wide llms.txt for each domain.
Each per-docset llms.txt covers all available versions of that product’s documentation.
llms.txt format
Each llms.txt file is a Markdown document structured as follows:
-
A top-level heading (
#) with the domain name. -
One or more second-level headings (
##) grouping pages, typically by product version for per-docset files, or by product for site-wide files. -
Under each heading, a bulleted list of links in standard Markdown format:
- [Page title](https://…page.md)
All URLs in llms.txt point directly to the .md markdown alternate for each page, with absolute URLs.
Example entries from docs.pingidentity.com/pingam/llms.txt:
# Documentation
## Pingam 8.1
- [/.well-known/webfinger](https://docs.pingidentity.com/pingam/8.1/am-oidc1/rest-api-oidc-discovery-webfinger.md)
- [AI agents](https://docs.pingidentity.com/pingam/8.1/am-oauth2/ai-agents.md)
- [Account lockout](https://docs.pingidentity.com/pingam/8.1/security/account-lockout.md)Recommended usage patterns
Loading context for a specific product
To give an agent focused context on a single product, point it at the product’s llms.txt:
https://docs.pingidentity.com/pingam/llms.txtThe agent can then fetch individual page Markdown alternates for the pages most relevant to the task.