noticed

search_people

Read-onlyIdempotent

Search people by free text — name, company, keywords. Known GitHub identities include github_login and github_url; github_user_id is a stable account ID, not a handle. Terms are OR-matched and ranked by how many match (more terms broadens, not narrows). To narrow: put AND between terms to require all (e.g. 'health AND medtech'), or prefix a term with + to require just it (e.g. '+rust berlin'). For roles/functions (founder, engineer, investor, …) use the role filter instead of free text — it catches title variants ('Founding Partner') that keywords miss, and free text over-matches bios/notes. To count people of a type, use structured filters and read total from the response — a free-text total counts keyword matches, not people of that type. scope:'own' (default) / scope:'public' (beyond your network + warm-intro paths). Optional company, location, skills, tags filters. A structured-filter zero = thin data, not absence — fall back to free text. Misspelled names/companies fall back to fuzzy matching (fuzzy: true = closest matches — confirm before trusting); concept queries with zero literal hits fall back to embedding similarity (semantic: true = related people, not literal matches). Free-text responses also return strong_total (rows matching ALL terms — the honest count) and per-row matched_on (which fields matched). In scope:'public', role/location/company/skills are applied to the global hits (filtered: true); filters that cannot apply there are listed in unsupported_filters.

Input Schema

TableJSON Schema

Name	Required	Description
`q`	No	Free-text query — matched OR-style against name, headline, company, location, skills. Operators narrow your own network: uppercase AND between terms requires all of them; a +term prefix requires just that term (e.g. '+health medtech'). In scope:'public', operators only affect the your-network half of results; global hits are matched by semantic similarity.
`role`	No	Filter by function (matched against headline): engineer, designer, product, gtm, founder, recruiter, investor, other. `gtm` covers sales/growth/marketing/partnerships/biz-dev; `other` is the catch-all for customer success, operations, legal, academia, etc. Matches any of the listed roles. Seniority (CEO, CTO, VP, etc.) is not a function — pass it in `q` instead.
`tags`	No	Require all of these tags you've applied.
`limit`	No	Page size (default 25, max 200 for own / 50 for public).
`scope`	No	'own' (default) — your network only. 'public' — global match + warm-intro paths to people you don't know yet. 'team' — your POOLED TEAM network (people known by ANY teammate); each hit carries `team_members` (who on the team knows them + their consent-gated strength).
`stale`	No	Only people you've never interacted with, or not in 90+ days.
`offset`	No	Pagination offset.
`recent`	No	Only people you've interacted with in the last 14 days.
`skills`	No	Require all of these skills.
`company`	No	Filter by company (matches current + former employers).
`sources`	No	Restrict to people who came from any of these sources.
`team_id`	No	For scope:'team' — the organizationId of the team to search (from account_status.teams). Optional when you're in exactly one team; required to pick when you're in more than one. A team you don't belong to is rejected.
`location`	No	Filter by location (substring match).
`added_after`	No	Only people added to your network ON/AFTER this date (ISO, e.g. 2026-06-01). For "who did I add this week / since X?". Adding ≠ interacting — for interaction recency use recent/stale.
`added_before`	No	Only people added STRICTLY BEFORE this date (ISO).
`last_interaction_after`	No	Only people whose last recorded touchpoint was ON/AFTER this date (ISO, e.g. 2026-06-22 or 2026-06-22T00:00:00Z). Use for 'who did I interact with on/since <date>' or 'everyone with a touchpoint this week'. People with no interaction are excluded. Combine with last_interaction_before for a date range.
`last_interaction_before`	No	Only people whose last recorded touchpoint was STRICTLY BEFORE this date (ISO, exclusive upper bound). People with no interaction are excluded. Combine with last_interaction_after for a date range.

Output Schema

TableJSON Schema

Name	Required	Description
`ok`	Yes	Whether noticed completed the operation.
`data`	No	The operation result when ok is true.
`error`	No	A human-readable error when ok is false.

Schema Changelog

Changes observed during successful MCP inspections.

Changed1 schema field changedAug 22, 2026

changedOutput schema / (root)

Previous value: -nullNew value: +{
+  "$schema": "http://json-schema.org/draft-07/schema#",
+  "additionalProperties": true,
+  "properties": {
+    "data": {
+      "additionalProperties": true,
+      "description": "The operation result when ok is true.",
+      "properties": {
+        "hasMore": {
+          "type": "boolean"
+        },
+        "hits": {
+          "items": {
+            "$ref": "#/properties/data/properties/people/items"
+          },
+          "type": "array"
+        },
+        "limit": {
+          "type": "number"
+        },
+        "offset": {
+          "type": "number"
+        },
+        "people": {
+          "items": {
+            "additionalProperties": true,
+            "properties": {
+              "company": {
+                "$ref": "#/properties/data/properties/people/items/properties/person_id"
+              },
+              "display_name": {
+                "type": "string"
+              },
+              "headline": {
+                "$ref": "#/properties/data/properties/people/items/properties/person_id"
+              },
+              "id": {
+                "type": "string"
+              },
+              "name": {
+                "type": "string"
+              },
+              "person_id": {
+                "type": [
+                  "string",
+                  "null"
+                ]
+              }
+            },
+            "type": "object"
+          },
+          "type": "array"
+        },
+        "scope": {
+          "enum": [
+            "own",
+            "team",
+            "public"
+          ],
+          "type": "string"
+        },
+        "strong_total": {
+          "type": "number"
+        },
+        "total": {
+          "type": "number"
+        }
+      },
+      "type": "object"
+    },
+    "error": {
+      "description": "A human-readable error when ok is false.",
+      "type": "string"
+    },
+    "ok": {
+      "description": "Whether noticed completed the operation.",
+      "type": "boolean"
+    }
+  },
+  "required": [
+    "ok"
+  ],
+  "type": "object"
+}

First observedAug 22, 2026

TDQS

A4.5/5.0

Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnlyHint=true and idempotentHint=true, so safety is covered, and the description adds substantial non-obvious behavior: OR-matching with broader results for more terms, ranked by match count, fuzzy fallback for misspellings, semantic/embedding fallback, and the caveat that free-text `total` counts keyword matches rather than people. It also exposes response nuances like `strong_total`, `matched_on`, `filtered`, and `unsupported_filters`.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Every sentence carries meaningful information and the main purpose is front-loaded, but the description is one dense paragraph with many interdependent clauses. It is rich enough to justify its length, yet the wall-of-text format makes it harder for an agent to scan than a bulleted structure would.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the tool's complexity, the description covers scoping (own/public), filtering fallbacks, count semantics, fuzzy/semantic behaviors, and public-mode filter limitations. An output schema exists, so return values need no detailed explanation; nothing an agent needs to invoke this tool correctly is missing.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so the baseline is 3, and the description adds real semantic value beyond it: OR/AND behavior, 'more terms broadens, not narrows', role-vs-free-text matching pitfalls, and the structured-zero caveat. However, it references `fuzzy: true` and `semantic: true` as if they were query options even though the schema does not define them and declares additionalProperties:false—a minor inconsistency that keeps it from a 5.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description opens with a specific verb+resource: 'Search people by free text — name, company, keywords.' It is clearly differentiated by its free-text search scope, but it never names sibling tools like get_person or list_intros, so the distinction is implicit rather than explicit.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Provides explicit when-to-use guidance: 'For roles/functions ... use the `role` filter instead of free text', 'To count people of a type, use structured filters and read `total` from the response', and 'A structured-filter zero = thin data, not absence — fall back to free text.' It also gives concrete narrowing syntax (AND, +prefix) and explains when public vs own scope is appropriate.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All noticed tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?