Read-only facts from brentburgess.com: profile, writing, projects and the site's own datasets.
Scored across 7 tools
Most tools have clearly distinct roles: fetch reads a specific document, search queries text, list_content enumerates content, and list_projects focuses on projects. There is some fuzzy overlap between search, list_content, and fetch for content discovery/retrieval, and get_site_stats versus query_dataset is related but distinguished as summary versus raw rows.
The set mostly follows a verb_noun pattern: get_profile, get_site_stats, list_content, list_projects, and query_dataset. The exceptions are fetch and search, which are bare verbs, but they remain readable and action-oriented.
Seven tools are well-scoped for a personal-site content server: discovery, retrieval, profile, projects, stats, and raw data each have a clear place. No tool appears redundant or out of scope.
The read-only surface covers search/discovery, full content retrieval, profile information, projects, site statistics, and underlying dataset queries. For a public site content server, there are no obvious gaps in the lifecycle of reading and exploring published material.
The full text of a post, case study, page or project as Markdown, or one section of it. The id can be an id from search or list_content (such as "work/sleep-anchor"), a path ("/about"), a brentburgess.com URL, a brentburgess:// URI or a unique slug ("consulting"). Add #anchor for one section, as search results do.
Inputs: id (required). Returns: id, url, text, title, +1 more
Who Brent Burgess is: his job title, a short introduction, where he is based, his current availability for work, his skills, how to reach him, and the main pages of the site. Use it first, and for any question about contacting or hiring him.
Inputs: no named parameters. Returns: url, name, intro, pages, +6 more
The figures on this site’s data page: browser page views and font downloads since launch, crawlers and scans in the latest complete week, commits, pull requests, CI runs and how long a merged change takes to go live, and the nightly pipeline’s tests and sources, each with the last day the data covers. For the tables behind them, use query_dataset.
Inputs: no named parameters. Returns: url, monthly, reading, building, +5 more
Every post, case study and page the server can read, newest first, with each one’s sections. Filter with kind ("post", "case_study" or "page"). Pass an id to fetch to read it.
Inputs: kind. Returns: count, items
The projects on the site: what each is, whether it is running, what it changed, the tools it uses, and its case study if there is one. Filter with status, for example "running".
Inputs: status. Returns: count, projects
The rows of one table behind this site’s data page, oldest first, with each column’s type and meaning. Filter by date with from and to (YYYY-MM-DD), which apply to the table’s date column; if more rows match than the limit, the most recent are returned. Daily and weekly tables keep about a year.
Inputs: dataset (required), to, from, limit. Returns: url, rows, grain, notes, +10 more
Full-text search over everything published: posts, case studies, the About, Consulting, Now, Privacy and Résumé pages (his work history is on the résumé), and the projects. Use plain keywords, such as "resumable upload" or "fixed-price audit". Each result is a section; pass its id to fetch to read it.
Inputs: query (required), kind, limit. Returns: query, total, results
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Read-only, sourced access facts for wealth tech products, vendor MCP servers and Form ADV columns.
Normalized official data with provenance, aggregations, insights, free samples and agent access.
Normalized official data with provenance, aggregations, insights, free samples and agent access.
Businesses in every country, each with dated facts and no ratings. Read-only.
Add one secure layer between your agents and this server.