MCP server
Tools
9 read-only tools. Start with get_profile, then list_case_studies or search_evidence, and cite the url on every record you use.
This page is generated from the server’s own tool list, so names, descriptions and inputs match what a client sees when it asks. To call a tool from code, see Connect from code.
Profile, status, and contact
Toolget_profileRead-only
Who Jason Sonderman is: headline, current availability, contact routes, the about page, and credentials. Call this first.
Inputs
None.
Example arguments
{}List case studies
Toollist_case_studiesRead-only
Every published case study with slug, title, one-line excerpt, role, timeline, tags, and the hiring lenses it belongs to. Optionally filter by lens or tag.
Inputs
| Name | Type | Required | Detail |
|---|---|---|---|
lens | string | No | Only case studies curated into this hiring lens. One of ic, leadership, fractional-consulting, peer. |
tag | string | No | Only case studies carrying this tag (case-insensitive). |
Example arguments
{
"lens": "ic"
}Get a case study
Toolget_case_studyRead-only
One case study by slug. `summary` returns structured fields: qaContext summary and facts, metrics, outcomes, team, contributions, and tools. `full` also returns the complete text as markdown, identical to the published .md mirror.
Inputs
| Name | Type | Required | Detail |
|---|---|---|---|
slug | string | Yes | Case-study slug, from list_case_studies. |
detail | string | No | One of summary, full. Default summary. |
Example arguments
{
"slug": "arcos-harmony-design-system",
"detail": "summary"
}List practice notes
Toollist_practice_notesRead-only
The shorter evidence tier: work that was part of the role rather than a project, or that shipped without a measured ending. Each note declares its evidence (documented, anecdotal, unfinished).
Inputs
None.
Example arguments
{}Get a practice note
Toolget_practice_noteRead-only
One practice note by slug, with its evidence tier and the full text. Do not supply a result or outcome the note does not state.
Inputs
| Name | Type | Required | Detail |
|---|---|---|---|
slug | string | Yes | Practice-note slug, from list_practice_notes. |
Example arguments
{
"slug": "arcos-feature-intake-gpt"
}Professional experience
Toolget_experienceRead-only
The professional experience record, most recent first: role, company, dates where shown, and bullets. This is the arbiter of chronology.
Inputs
None.
Example arguments
{}Colleague attestations
Toolget_attestationsRead-only
Verbatim quotes from people who worked with Jason: the only content on the site he did not write. Attribute every quote; never restate one in Jason’s voice or convert it into a metric.
Inputs
None.
Example arguments
{}Curated reading order for a hiring decision
Toolget_hiring_lensRead-only
The case studies Jason would point you to first when screening for a given kind of role, in order, with a one-line reason for each.
Inputs
| Name | Type | Required | Detail |
|---|---|---|---|
lens | string | Yes | ic: hands-on individual-contributor design role. leadership: design leadership. fractional-consulting: a fractional or consulting engagement. One of ic, leadership, fractional-consulting. |
Example arguments
{
"lens": "leadership"
}Search the evidence
Toolsearch_evidenceRead-only
Keyword search across case-study facts, metrics, outcomes, and summaries, practice notes, and attestation quotes. Returns short verbatim passages with their source, evidence tier, and URL. Deterministic, not semantic: try specific terms (e.g. “design tokens”, “field research”, “hiring”) and a few variations.
Inputs
| Name | Type | Required | Detail |
|---|---|---|---|
query | string | Yes | 1 to 300 characters. |
limit | integer | No | 1 to 25. Default 8. |
Example arguments
{
"query": "design tokens",
"limit": 5
}