Perplexity Launches Computer for Counsel: A Multi-Model Agentic Layer for Legal Workflows

Illustrative simulation for engineers. Not connected to live legal data or real client matters.</p>
</div>

<div class=”section-label”>1 · Choose a legal workflow</div>
<div class=”tabs” id=”tabs”>
<div class=”tab active” data-k=”nda”>
<div class=”t”>Third-party NDA intake</div>
<div class=”d”>Review red flags, fill entities, route for signature.</div>
</div>
<div class=”tab” data-k=”reg”>
<div class=”t”>Regulatory monitoring</div>
<div class=”d”>Build a dashboard of US state privacy & adtech laws.</div>
</div>
<div class=”tab” data-k=”case”>
<div class=”t”>Case research + citations</div>
<div class=”d”>Research non-compete enforceability, export cited PDF.</div>
</div>
</div>

<div class=”controls”>
<button class=”act run” id=”run”> Run agentic workflow</button>
<button class=”act reset” id=”reset”>Reset</button>
</div>

<div class=”progress”><div class=”bar” id=”bar”></div></div>
<div class=”section-label”>2 · Pipeline execution</div>
<div class=”pipe” id=”pipe”></div>

<div class=”out” id=”out”>
<h3 id=”outTitle”>Deliverable</h3>
<div class=”deliver” id=”outBody”></div>
</div>

<div class=”foot”>
<span class=”src”>Facts sourced from Perplexity’s “Introducing Computer for Counsel” (Jun 24, 2026).</span>
<span class=”brand”>Marktechpost</span>
</div>
</div>

<script>
// —- Illustrative workflow definitions (facts drawn from Perplexity’s launch blog) —-
var FLOWS = {
nda: {
title: “Clean NDA package, routed for signature”,
steps: [
{n:”Plan task”, model:null, meta:”planner”, body:”Decompose into: extract parties → flag risky clauses → normalize template → route for e-signature.”},
{n:”Route models”, model:[“Reasoning model”,”Extraction model”], meta:”router · 20+ models”, body:”Router assigns a reasoning model to clause analysis and an extraction model to entity capture.”},
{n:”Call MCP connectors”, model:null, conn:[“Docusign”,”NetDocuments”], meta:”MCP”, body:”Pulls the counterparty draft from NetDocuments; prepares the Docusign envelope and signatory fields.”},
{n:”Synthesize”, model:[“Reasoning model”], meta:”synthesis”, body:”Flags an uncapped indemnity and a missing residuals clause. Fills entity and signatory information. Prepares a clean copy.”},
{n:”Verify & route”, model:null, meta:”human-in-loop”, body:”Each flag links to the source clause. Lawyer approves; Computer routes for signature via Docusign.”}
],
deliver: ‘Reviewed NDA with 2 red flags surfaced <a class=”cite” title=”Illustrative source link”>[clause 4.2]</a> <a class=”cite” title=”Illustrative source link”>[clause 7.1]</a>, entities auto-filled, and a clean copy queued in Docusign for approval. The attorney makes the final call before anything is sent.’
},
reg: {
title: “Shareable US state privacy & adtech dashboard”,
steps: [
{n:”Plan task”, model:null, meta:”planner”, body:”Decompose into: enumerate states → check law status → gather cases → build shareable dashboard.”},
{n:”Route models”, model:[“Research model”,”Reasoning model”], meta:”router · 20+ models”, body:”Router uses a research model for jurisdiction sweep and a reasoning model for status classification.”},
{n:”Call MCP connectors”, model:null, conn:[“Midpage”,”Open web”], meta:”MCP”, body:”Queries Midpage for statutes and relevant cases; cross-checks effective dates on the open web.”},
{n:”Synthesize”, model:[“Reasoning model”], meta:”synthesis”, body:”Marks which states have laws in effect vs pending. Cites Midpage for each holding.”},
{n:”Publish dashboard”, model:null, meta:”export”, body:”Generates a shareable dashboard. Every status cell links to its statute or case.”}
],
deliver: ‘A shareable dashboard showing which US states have privacy/adtech laws in effect vs pending, each cell citing Midpage <a class=”cite” title=”Illustrative source link”>[statute]</a> <a class=”cite” title=”Illustrative source link”>[case]</a>. Built to be re-run as laws change.’
},
case: {
title: “Non-compete enforceability memo (PDF, cited)”,
steps: [
{n:”Plan task”, model:null, meta:”planner”, body:”Decompose into: scope question → retrieve cases → run citator → summarize → export PDF.”},
{n:”Route models”, model:[“Research model”,”Reasoning model”], meta:”router · 20+ models”, body:”Research model gathers post-2024 FTC-ban cases; reasoning model weighs enforceability.”},
{n:”Call MCP connectors”, model:null, conn:[“Midpage citator”], meta:”MCP”, body:”Midpage citator checks whether each cited case is still good law.”},
{n:”Synthesize”, model:[“Reasoning model”], meta:”synthesis”, body:”Summarizes key cases. Flags unsettled splits across circuits. Notes where law is in flux.”},
{n:”Export PDF”, model:null, meta:”export”, body:”Exports a PDF memo. Every claim hyperlinks to its underlying authority.”}
],
deliver: ‘A PDF memo on non-compete enforceability after the FTCu2019s 2024 ban: key cases summarized, unsettled splits flagged <a class=”cite” title=”Illustrative source link”>[case A]</a> <a class=”cite” title=”Illustrative source link”>[case B]</a>, each verified by the Midpage citator before export.’
}
};

var active = “nda”, running = false, timers = [];

function postHeight(){
try{
var h = document.getElementById(‘root’).offsetHeight + 40;
parent.postMessage({type:’cfc-resize’, height:h}, ‘*’);
}catch(e){}
}

function clearTimers(){ timers.forEach(clearTimeout); timers = []; }

function buildSteps(){
var pipe = document.getElementById(‘pipe’);
pipe.innerHTML = ”;
FLOWS[active].steps.forEach(function(s,i){
var el = document.createElement(‘div’);
el.className = ‘step’; el.id = ‘step’+i;
var chips = ”;
if(s.model){ s.model.forEach(function(m){ chips += ‘<span class=”chip model”>’+m+'</span>’; }); }
if(s.conn){ s.conn.forEach(function(c){ chips += ‘<span class=”chip”>’+c+'</span>’; }); }
el.innerHTML =
‘<div class=”row”><span class=”dot”></span><span class=”name”>’+(i+1)+’. ‘+s.n+'</span>’+
‘<span class=”meta”>’+s.meta+'</span></div>’+
‘<div class=”body”>’+s.body+'</div>’+
(chips?(‘<div class=”chips”>’+chips+'</div>’):”);
pipe.appendChild(el);
});
}

function setTab(k){
if(running) return;
active = k;
document.querySelectorAll(‘.tab’).forEach(function(t){
t.classList.toggle(‘active’, t.dataset.k===k);
});
reset();
}

function reset(){
clearTimers(); running = false;
document.getElementById(‘run’).disabled = false;
document.getElementById(‘bar’).style.width = ‘0’;
document.getElementById(‘out’).classList.remove(‘show’);
buildSteps();
setTimeout(postHeight, 50);
}

function run(){
if(running) return;
running = true;
document.getElementById(‘run’).disabled = true;
document.getElementById(‘out’).classList.remove(‘show’);
var steps = FLOWS[active].steps;
var n = steps.length;
steps.forEach(function(s,i){
// reveal
timers.push(setTimeout(function(){
var el = document.getElementById(‘step’+i);
el.classList.add(‘show’,’running’);
document.getElementById(‘bar’).style.width = ((i+0.5)/n*100)+’%’;
postHeight();
}, i*900 + 150));
// complete
timers.push(setTimeout(function(){
var el = document.getElementById(‘step’+i);
el.classList.remove(‘running’); el.classList.add(‘done’);
document.getElementById(‘bar’).style.width = ((i+1)/n*100)+’%’;
postHeight();
}, i*900 + 750));
});
// final output
timers.push(setTimeout(function(){
document.getElementById(‘outTitle’).textContent = ‘✓ ‘ + FLOWS[active].title;
document.getElementById(‘outBody’).innerHTML = FLOWS[active].deliver +
‘<br><br><span style=”color:#76B900;font-size:11px;”>Judgment and strategy stay with the lawyer. Enterprise data is not used for training.</span>’;
document.getElementById(‘out’).classList.add(‘show’);
document.getElementById(‘run’).disabled = false;
running = false;
postHeight();
}, n*900 + 400));
}

document.getElementById(‘tabs’).addEventListener(‘click’, function(e){
var t = e.target.closest(‘.tab’); if(t) setTab(t.dataset.k);
});
document.getElementById(‘run’).addEventListener(‘click’, run);
document.getElementById(‘reset’).addEventListener(‘click’, reset);

buildSteps();
window.addEventListener(‘load’, function(){ setTimeout(postHeight, 80); });
window.addEventListener(‘resize’, postHeight);
setTimeout(postHeight, 200);
</script>
</body>
</html>
”
style=”width:100%;border:0;display:block;height:600px;overflow:hidden;”
loading=”lazy” title=”Computer for Counsel — Agentic Workflow Simulator”>

# Perplexity's cited search primitive — the foundation Computer for Counsel builds on.
# This is the public Sonar API, not the Computer for Counsel product itself.
import os, requests

resp = requests.post(
    "https://api.perplexity.ai/chat/completions",
    headers={"Authorization": f"Bearer {os.environ['PERPLEXITY_API_KEY']}"},
    json={
        "model": "sonar-pro",
        "messages": [
            {"role": "user",
             "content": "Is a non-compete signed in California enforceable in 2026?"}
        ],
        # Restrict grounding to trusted sources, the way a lawyer would.
        "search_domain_filter": ["law.cornell.edu", "courtlistener.com", "ca.gov"],
    },
    timeout=60,
)
data = resp.json()
print(data["choices"][0]["message"]["content"])   # the answer
for url in data.get("citations", []):              # every source, for verification
    print("source:", url)