Skip to content
CLMSpace
The roster

The twelve agents behind the platform.

Each agent is a scoped capability with one job — read the PDF, hold the row until a human verifies it, compare the clause to the scoped playbook, draft the counter-position from your clause library, assemble the renewal pack. Priced individually, packaged into four tiers. Click any agent to jump to the deep dive.

Clause + obligation parser

Extractor

Turns every contract — PDF, Word, Markdown — into structured obligations.

Extractor reads contracts in PDF, DOCX, MD, or plain text using PyMuPDF for layout + python-docx for Word. Claude Haiku does the structured extraction via tool_use with a strict JSON schema — no free-text wandering. Every obligation arrives with a verbatim source quote, page number, and (for PDFs) a bounding box you can click to view the source clause in place.

What it delivers
  • PDF / DOCX / MD / TXT support
  • Tool-use extraction with guaranteed schema adherence
  • Per-obligation citation: verbatim quote + page + bbox
  • Handles 20+ page MSAs with chunked extraction + streaming

From £600/month à-la-carte · included in the CLM Suite tier and above.

ExtractorAgent
  • PDF / DOCX / MD / TXT support
  • Tool-use extraction with guaranteed schema adherence
  • Per-obligation citation: verbatim quote + page + bbox

Human-in-the-loop reviewer

Verifier

Routes every AI-extracted obligation to a human before it binds anything.

Verifier maintains the AI / Human-Verified / Human-Overridden lifecycle state on every obligation. Until an admin signs off, the obligation is excluded by default from the ERP payment-schedule export, the CRM counterparty snapshot, and the recommended renewal action. Review-queue ranking surfaces high-value old extractions first.

What it delivers
  • Three-state verification: AI_Extracted → Verified / Overridden
  • Review queue with value × age priority
  • Per-binding ERP/CRM gate — unverified rows excluded
  • Audit log of every edit with reason + actor + timestamp
  • Unverified-aged alert tier (default 14 days)

From £400/month à-la-carte · included in the CLM Suite tier and above.

VerifierAgent
  • Three-state verification: AI_Extracted → Verified / Overridden
  • Review queue with value × age priority
  • Per-binding ERP/CRM gate — unverified rows excluded

Standard-vs-actual auditor

Deviations

Flags every contract clause that departs from your approved standards.

Deviations consumes the resolved standard positions (Tier 1 templates > Tier 2 verified history > manual config) and compares every extracted obligation against the right tier. Each departure gets a risk_level + standard + actual + recommendation, persisted on the obligation. High-risk findings open an approval card in Teams; medium/low feed the dashboard's risk tile.

What it delivers
  • Tier-aware comparison (Tier 1 / Tier 2 / Manual)
  • Risk-tiered findings (High / Medium / Low)
  • Teams adaptive-card escalation for High
  • Approval workflow with audit trail

From £700/month à-la-carte · included in the Foundation tier and above.

DeviationsAgent
  • Tier-aware comparison (Tier 1 / Tier 2 / Manual)
  • Risk-tiered findings (High / Medium / Low)
  • Teams adaptive-card escalation for High

Counter-position drafter

Negotiator

Generates redlines anchored to your clause library — never invented prose.

Negotiator reads the obligation, the resolved standard for its type, and the preferred / fallback / walk-away clauses from your clause library. Claude is constrained to reuse library text verbatim in the redline; it never invents contractual language that conflicts with what you've approved. Output ships as a copyable Markdown clause + a paragraph for the reply email.

What it delivers
  • Per-deviation negotiation pack
  • Suggested redline (Markdown, copyable)
  • Sample email response paragraph
  • Preferred / fallback / walk-away clause callouts

From £950/month à-la-carte · included in the Commercial Office tier and above.

NegotiatorAgent
  • Per-deviation negotiation pack
  • Suggested redline (Markdown, copyable)
  • Sample email response paragraph

Renewal / amendment / termination drafter

Drafter

Drafts the document — informed by every prior amendment with this counterparty.

Drafter pulls the counterparty's full agreement chain (MSA + SOWs + variations), resolves your active standard positions, and produces a ready-to-edit Markdown draft. Departures from the standard are flagged inline with a [DEVIATION] marker. Unknowns surface as [TO CONFIRM: …] placeholders rather than guesses.

What it delivers
  • 5 draft types: renewal / non-renewal / amendment / termination / brief
  • Grounded in counterparty consolidated view
  • deviations_from_standard JSON block on every draft
  • [TO CONFIRM: …] placeholders for unknowns

From £800/month à-la-carte · included in the Foundation tier and above.

DrafterAgent
  • 5 draft types: renewal / non-renewal / amendment / termination / brief
  • Grounded in counterparty consolidated view
  • deviations_from_standard JSON block on every draft

Lifecycle orchestrator

Renewals

Builds a one-shot renewal pack 60–120 days before every expiry.

Renewals watches expiry dates across the portfolio and assembles a decision pack for each contract in the window: key facts, risk score, standard-vs-actual delta per obligation type, recommended action with rationale, and a Markdown draft auto-typed to the recommendation. One click, one decision, every input staged.

What it delivers
  • Upcoming renewals list with days-to-expiry urgency
  • Per-contract renewal pack with recommended action
  • Standard-vs-actual table with tier provenance
  • Auto-typed draft (renewal / amendment / non-renewal / brief)

From £550/month à-la-carte · included in the Foundation tier and above.

RenewalsAgent
  • Upcoming renewals list with days-to-expiry urgency
  • Per-contract renewal pack with recommended action
  • Standard-vs-actual table with tier provenance

Counterparty risk scorer

Risk

Composite 0–100 score across deviations, exposure, verification ratio, and age.

Risk computes a transparent attention score per counterparty: open high/medium deviations, unverified obligations ratio, imminent expiry, portfolio concentration (>25% of annual value), and stale-review days. Every point in the score traces back to a reason chip the user can click into. Surfaces on the counterparty list, dashboard, and renewal pack.

What it delivers
  • Composite 0–100 score with reason breakdown
  • Bands: Low / Medium / High
  • Portfolio-wide ranking for triage
  • Drives recommended action in the renewal pack

From £450/month à-la-carte · included in the Commercial Office tier and above.

RiskAgent
  • Composite 0–100 score with reason breakdown
  • Bands: Low / Medium / High
  • Portfolio-wide ranking for triage

Tiered standards keeper

Standards Learner

Derives your negotiation baseline from your own template drafts and verified history.

Drop a template MSA / NDA / SOW into the bound SharePoint folder; the agent extracts its obligations, tags them as template-derived, and they become Tier 1. Verified obligations on executed contracts feed Tier 2 via statistical aggregation per (scope × obligation_type) pair — median value, modal currency / trigger / unit; the signal sharpens with more verified samples. Manual fallback covers the rest. Standards admin tab shows the tier badge on every active position.

What it delivers
  • Auto-derives Tier 1 from SharePoint /Templates
  • Mines Tier 2 from verified historical obligations
  • Tier provenance on every active standard
  • Admin editor for the manual fallback

From £350/month à-la-carte · included in the Commercial Office tier and above.

Standards LearnerAgent
  • Auto-derives Tier 1 from SharePoint /Templates
  • Mines Tier 2 from verified historical obligations
  • Tier provenance on every active standard

SharePoint folder ingest

Watcher

Bind a folder; new files become structured obligations within 5 minutes.

Watcher uses Microsoft Graph's app-only token (Sites.Read.All + Files.Read.All) to walk bound folders on a 5-minute cron. The first poll establishes a baseline cursor; subsequent polls pick up only net-new arrivals. Each ingest streams through the Extractor + Verifier pipeline. Live progress per binding in the Settings tab.

What it delivers
  • Per-tenant bound folders (Templates / Agreements roles)
  • Cursor-based ingest — never misses, never re-ingests
  • On-bind backfill of existing files
  • Live per-file progress with graceful handling of Anthropic 429s

From £0/month à-la-carte · included in the Foundation tier and above.

WatcherAgent
  • Per-tenant bound folders (Templates / Agreements roles)
  • Cursor-based ingest — never misses, never re-ingests
  • On-bind backfill of existing files

Conversational portfolio queryer

Ask CLM

Natural-language questions over your obligation graph, citation-grounded.

Ask CLM sends the relevant slice of your obligation + agreement graph to Claude as context and gets back grounded answers with inline obligation-ID references rendered as hoverable, clickable chips. Persisted chat history per user, soft-deletable with tenant audit. Available from any view via a right-slide panel (Cmd/Ctrl + /).

What it delivers
  • Natural-language Q&A over contracts + obligations
  • Citation chips with hover preview + click-through to grid
  • Persisted chat history per user, soft-delete + audit
  • Markdown + tables rendered properly

From £400/month à-la-carte · included in the Foundation tier and above.

Ask CLMAgent
  • Natural-language Q&A over contracts + obligations
  • Citation chips with hover preview + click-through to grid
  • Persisted chat history per user, soft-delete + audit

Clause library keeper

Curator

The approved clause language your negotiators reuse.

Curator stores per-tenant approved clause text keyed by obligation type with a preference tier (preferred / fallback / walk_away). Admins edit through the Clauses tab. The Derive button seeds the library from your Tier-1 templates by pulling the longest verbatim description per type; you then refine.

What it delivers
  • Per-type clause library with preference tiers
  • Derive-from-templates button (Tier-1 seed)
  • Admin CRUD via Clauses tab
  • Powers Negotiator + Drafter redline text

From £250/month à-la-carte · included in the Commercial Office tier and above.

CuratorAgent
  • Per-type clause library with preference tiers
  • Derive-from-templates button (Tier-1 seed)
  • Admin CRUD via Clauses tab

Posture + event recorder

Lifecycle

The system flags, the user decides. Never silently auto-decides.

Lifecycle records explicit events: Confirm Active, Renewed, Terminated, Expired, Suspended, Reactivated. Indefinite-term agreements with no recent event surface as 'Unverified — needs review' rather than being assumed Active or Expired. Alert tiers fire at 90 / 60 / 30 / 14 days for obligations, agreement expiries, and stale reviews.

What it delivers
  • Explicit lifecycle event recording with audit trail
  • 'Unverified — needs review' first-class posture
  • 90 / 60 / 30 / 14 day alert tiers
  • Daily alert engine with Teams digest

From £0/month à-la-carte · included in the CLM Suite tier and above.

LifecycleAgent
  • Explicit lifecycle event recording with audit trail
  • 'Unverified — needs review' first-class posture
  • 90 / 60 / 30 / 14 day alert tiers

Not sure which agents fit your stack?

30-minute discovery call: we walk through your current contract stack (SharePoint, Microsoft 365, any existing CLM), your renewals cadence, and your negotiation playbook gaps — then recommend a tier or bespoke agent mix. We quote on the call.