MCP App Store
by KeenEthics
Intelligro icon

Intelligro MCP Server for ChatGPT

By Intelligro

Overview

Intelligro is a workforce decision intelligence platform. It helps leaders determine the right mix of human, AI, external, and organizational capability required to achieve a business outcome. Your systems hold the data. Intelligro helps decide what to do with it. Tell it the outcome you need, such as growing revenue 20% this year, keeping service levels while volume doubles, entering a new market, or cutting a process from 12 days to 5. Intelligro compares nine ways to get there: hire, develop, redeploy, automate, augment with AI, borrow, partner, restructure, and acquire. It recommends a plan against your timeline, budget, and constraints, shows the assumptions and benchmarks behind it, and asks only the questions that would change the answer. From there you can test scenarios, generate an execution plan, record what you decided, and track checkpoints and results. Every decision teaches the system what works in your organization, making future recommendations more relevant and accurate. Built for where leaders already work. Intelligro works above your existing systems (HRIS, finance, ATS, ERP), not as another system of record. Intelligro uses group-level data only (groups of 5 or more), never evaluates individual employees, and a person reviews every recommendation before any decision about people is made.

How to connect

  1. ChatGPT: Open the Intelligro listing in the ChatGPT app directory and choose Connect. ChatGPT listing Intelligro exposes 13 tools here.

Tools

analyze_intervention

ChatGPT
Engine action: the engine validates the request and records it. Analyzes how an intervention path would work for a finished decision, inside the decision object. Built: "Augment with AI" (how AI changes the work, the role, the human capabilities needed and the transition: work allocation, capacity, role evolution, future capabilities, human transition, paired interventions, implementation requirements, checkpoints and expected outcomes). The engine computes every score with versioned models; never compute or adjust a score yourself. AI-led candidate work is not an automation recommendation and FTE-equivalent capacity is not jobs eliminated: say so when you relay them. Group level only: roles, tasks and cohorts (5 or more people), never named people. Mark each input's source_type ("current organization fact" for what the user stated, "model inference" for anything you proposed). Ask the user the questions in missing_inputs, as written. Scenarios (mode "scenario") never change the baseline unless saved and then selected.

export_brief_pdf

ChatGPT
Read only. Creates a PDF of a finished brief, with its milestone checklist and the human-review notice, and returns a download_url that works for 30 minutes.

generate_brief

ChatGPT
Engine action: the engine validates the request and records it. Writes the execution brief for a finished decision from run_decision: with its decision_id, the brief is written for each path in the recommended plan, from the decision's facts, assumptions and checkpoints. Returns a job_id right away; the brief takes about 30 to 110 seconds. Get it with poll_brief_result.

get_connected_team_data

ChatGPT
Read only. Returns the organization's latest headcount and team structure from its connected systems: group-level cohorts (groups of 5 or more) and spans and layers, with the source and when it was pulled. Offer to use it in a decision, and pass the cohorts to run_decision with the source.

get_data_templates

ChatGPT
Read only. Returns the group-level workforce data Intelligro can use, what it never accepts (names, emails, IDs, individual pay or ratings), and download links for the team cohorts and spans and layers templates.

get_decision_memory

ChatGPT
Read only. Returns what the organization has learned from its earlier decisions: learnings with their confidence, interventions it ruled out before and why, and recent decisions.

log_outcome

ChatGPT
Engine action: the engine validates the request and records it. Records what the user decided about a finished recommendation (Followed, Modified, Overridden or Deferred) in the organization's decision history, with who decided (a role, not a name). Sets the check-in and outcome-close dates from the time horizon.

poll_brief_result

ChatGPT
Read only. Returns the status of a brief started with generate_brief. "processing" is normal: check again in 10 to 15 seconds. "done" returns the brief, its root cause fields, milestones, the figures check (which traces every number to the user's inputs or cited research) and the final decision_confidence, plus a compliance_notice for the user.

record_milestone

ChatGPT
Engine action: the engine validates the request and records it. Records an expected-versus-actual checkpoint for a logged decision. A difference of 10% or more, with the reason, is saved as a learning for the organization's future decisions. Recording the same checkpoint again replaces its earlier value. Group-level measures only.

record_outcome

ChatGPT
Engine action: the engine validates the request and records it. Records an observed result for a logged decision on one of four clocks: Implementation, Workforce / Operational, Business or Financial. Met targets on followed recommendations, and financial estimates that missed by 10% or more, are saved as learnings. Group-level measures only.

run_decision

ChatGPT
Engine action: the engine validates the request and records it. Runs a full Intelligro decision from the business outcome and everything the user has said: research, scoring of all 9 workforce paths, the sufficiency check, the outcome model (recruiting throughput or capacity) and the math across the unknowns. Never asks questions: anything not stated is filled from benchmarks or labeled defaults and listed as an assumption to check, ranked by how much it could change the answer. Returns the recommendation, plans, assumptions, checkpoints and a decision_id, and shows the Intelligro decision card. If a fact the user can know today would change the answer, it returns status "needs_input" with at most 2 questions instead of a recommendation. Uses one of the organization's monthly decisions. If status is "running", call again with only the decision_id. Send the user's question as they asked it. Never reword a question about one specific person (fire, discipline, promote, pay or keep a named or described individual) into an organization question: the engine declines those, and you relay its explanation.

save_scenario

ChatGPT
Engine action: the engine validates the request and records it. Saves a scenario for a logged decision: the scenario model, its inputs and outputs, the limiting factor, and whether it is the chosen one. Passing an existing scenario_id replaces that scenario.

update_decision

ChatGPT
Engine action: the engine validates the request and records it. Use whenever the user changes, corrects or asks "what if" about any number or fact in a decision from run_decision (team size, volume, timeline, budget, a constraint); do not answer those from your own arithmetic. Applies the correction and returns the re-scored result with a "what_changed" line. Numbers go in changes (keys from the assumptions list) and are re-scored in code in seconds. A new fact or limit in words goes in new_facts; if it could change the judgment, the paths are re-scored too. Does not use another decision.

Capabilities

WritesInteractive

Example Prompts

Click any prompt to copy it.

Similar Apps

Listed on MCP App Store

Listed on MCP App Store

Add this badge to your site or README. It links to this page.

HTML

<a href="https://mcp-app-store.keenethics-labs.com/app/intelligro"><img src="https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg" alt="Listed on MCP App Store" width="220" height="44"></a>

Markdown

[![Listed on MCP App Store](https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg)](https://mcp-app-store.keenethics-labs.com/app/intelligro)