Skip to content
fidelic.ai®®

Expert tier · Legal / Compliance

PRAX-01

AI Contract Review

I redline contracts against the playbook your firm has approved. I surface deviations and suggest fallbacks. The decision to ship the redline stays with the supervising attorney.

PRAX-01, in her own words

Scope the role first. Deploy only after approval.

At a glance

Tier
Expert · a small fraction of comparable senior salary
Reports to
A supervising attorney
Primary work
Contract redlining + playbook deviation flagging
Will not do
Finalize redlines, change playbook, ship without attorney review
Success criterion
Playbook-deviation detection rate + redline-acceptance rate

About this role

PRAX redlines contracts against your firm's playbook — surfacing deviations, suggesting fallback positions, flagging the clauses that require senior attorney review.

Spellbook lists at ~$149–199/user/mo with sales-gated pricing and Word add-in delivery. PRAX is flat monthly per supervising attorney, with the playbook frozen at deployment and redlining gated by review.

Areas of focus

  • Redlines contracts against your firm's published playbook
  • Surfaces every deviation from playbook positions with the suggested fallback
  • Flags clauses that require senior attorney review (indemnity caps, liability, IP)
  • Maintains a per-counterparty redline log
  • Logs every redline with the playbook reference and deviation type
Where I push hardest

PRAX will not finalize a redline. Every output is a draft for the supervising attorney's review, with the playbook reference and deviation type visible per redline.

What surprises new clients

Every Friday PRAX ships a contract-volume digest: which counterparties most often demand playbook deviations, which clauses get redlined the most, where the firm's positions soften over time. Negotiation intelligence, on the record.

My stack

Tools I use

SlackWordGoogle DocsIroncladDocuSign

Background

Where I come from
Contract review is the legal function with the highest contract-volume-to-attorney-headcount ratio in most firms. PRAX is built to pre-process redlines, not to replace attorney judgment. Every output is a draft for review.
How I think about the work
  • Reads the firm's playbook and last 90 days of redline history before working a contract
  • Routes against the four-tier constitution: autonomous on first-pass redlines, review-required on senior-flagged clauses, escalate on negotiation questions, refuse on playbook updates
  • Logs every redline with the playbook reference and deviation type
How I've been tested
EvalOps suite covers playbook-deviation detection accuracy, redline precision against attorney-approved baselines, and false-flag rate.
Where I'm running today
First-cohort deployments scheduled May–June 2026.

What I won't take on

PRAX will not finalize a redline; every output requires supervising-attorney review.

PRAX will not change the firm's playbook. The playbook is updated by the supervising attorney, not the agent.

PRAX will not negotiate directly with counterparties; that authority stays with humans.

At the floor, not the average

PRAX flags low confidence on ambiguous clauses rather than emitting a confident redline. The supervising attorney decides.

The first 30 days

  1. Day 1

    Provisioned. Supervising attorney uploads the playbook and approves constitutional rules.

  2. Week 1

    First contract redlines under review. Playbook-deviation detection baseline established.

  3. Month 1

    Per-counterparty redline log compounding. Negotiation patterns surfaced weekly.

What success looks like at 30 days

Playbook-deviation detection rate at the level your supervising attorney defines, with redline-acceptance rate above the agreed threshold for three weeks.

What I'll need from you

Document storage (Word, Google Docs, Ironclad). Slack for digests. Playbook uploaded by supervising attorney at deployment.

Engagement

Expert tiera small fraction of a junior attorney (contract-review-only scope) salary

Junior attorney (contract-review scope): $9.2–14.6K/mo fully loaded (NALP 2025). PRAX: a small fraction of the comparable salary. Spellbook: ~$149–199/user/mo with sales-gated pricing.

PRAX-01 costs a small fraction of what a junior attorney (contract-review-only scope) costs. A junior attorney (contract-review-only scope) runs $9–15K/month fully loaded, and we don’t price against that — PRAX-01 doesn’t do what the person in that role does. PRAX-01 does the daily work that should already be in your inbox by Monday morning: the briefings, the structured first drafts, the early-warning monitors, the analysis that surfaces the question worth thinking about. That person — a real human, on your team — owns the work a fidelic agent can’t take on: the unfamiliar judgment, the customer in the room, the call that needs a name on it. You can keep both. That’s the point. Agency hiring speed, without the agency price. See the math on /pricing.

Terms

  • Cancel any month with 30 days' notice
  • All redlines reviewed by the supervising attorney before ship
  • Playbook is frozen at deployment; only the attorney updates it
  • EvalOps suite gates every release
  • Negotiation authority stays with humans

What you actually get

How it works

You see exactly what the agent will do — day one, week one, month one — before you pay anything.

First minutes
A short voice call walks through what you need. You get three agent options. Connect Slack. Your agent is live in your team chat.
Day 1
The agent reads what you point it to — Slack channels, docs, customer notes. It asks you questions in DMs when it doesn't know something. No pretending.
Week 1
First real work shows up for you to review — a brief, a draft, a triage report. You sign off on what's good and flag what isn't. The agent adjusts.
Month 1
The role is up and running. Your agent knows when to loop you in. The one number you said you'd measure has its first reading.

Security model

How a fidelic agent runs

  • Each customer deployment runs in an isolated Anthropic project.
  • Agents only see the Slack channels and docs you give them access to.
  • We log what the agent did, not what was said in your channels or files.
  • Every agent has clear rules for what it can do on its own — and what needs you to sign off.

Read the full security model →

The line we don’t cross

What humans still own

Fidelic agents do not replace human judgment in unfamiliar, political, relational, or high-stakes situations. The agent handles the repeatable work around those decisions so the human can move faster.

  • Final approval on strategic accounts.
  • Budget, refunds, policy, legal, and hiring decisions.
  • Customer relationships and any sensitive escalation.
  • Any action above the agent’s written authority.

Pairs well with

Related Hard Questions

PRAX-01 — AI Contract Review (alternative to Spellbook) · FidelicAI