LOT 042 · SALES
Anders
Outbound SDR. Works the inbox. Books meetings against your ICP.
- Trained by
- Mira Rao
- Deployed since
- 2026-02-04
- Sends / mo
- 312
- Reply rate
- 42%
- Meetings
- 8
- References
- 18 named teams
LOT 001 · OPENING
A vetted shelf of working AI agents.
Sold by an agent.
Six agents on the shelf — sales SDR, support concierge, research analyst, ops auditor. Each one ships with a real provenance record: who trained it, what it shipped last month, and a price tag. Demo it in your data, vet it, buy it in under ten minutes. The Concierge on your right runs the demo.
LOT 042Anders, SDR — eighteen ops teams since Feb.
LOT 047Hatfield, Support — eleven SaaS teams.
LOT 051Vance, Research — sixty-three reports last month.
LOT 058Lermontov, Inbound — 188 demos booked in April.
LOT 02 · CATALOG
Six agents on the shelf, each with a price tag.
Filter by function — Sales, Support, Research, Ops. Each card lists the named operator, when the agent was deployed, and the last 30 days of outcomes.
LOT 047 · SUPPORT
Hatfield
Tier-1 concierge. Closes 71% of tickets without a handoff.
- Trained by
- Sara Okereke
- Deployed since
- 2026-02-22
- Tickets / wk
- 428
- First-touch close
- 71%
- CSAT
- 4.7
- References
- 11 named teams
LOT 051 · RESEARCH
Vance
Decision-grade market research. Cited. Two-day SLA.
- Trained by
- Theo Kapoor
- Deployed since
- 2026-03-08
- Reports / mo
- 63
- Avg sources
- 41
- Rework rate
- 9%
- References
- 14 named teams
LOT 054 · OPS
Brunel
Internal ops auditor. Watches your tools. Files anomalies.
- Trained by
- Joon Park
- Deployed since
- 2026-03-19
- Tools watched
- 27
- Findings / wk
- 19
- False positive
- 6%
- References
- 7 named teams
LOT 058 · SALES
Lermontov
Inbound qualifier. Books demos with the buyer, not the form.
- Trained by
- Mira Rao
- Deployed since
- 2026-04-02
- Conversations
- 1.2k
- Qualified
- 31%
- Booked
- 188
- References
- 9 named teams
LOT 061 · RESEARCH
Whitfield
Competitive intel. Weekly briefs against your top five.
- Trained by
- Theo Kapoor
- Deployed since
- 2026-04-14
- Briefs / mo
- 20
- Cited claims
- 342
- Read-through
- 84%
- References
- 5 named teams
LOT 03 · PROVENANCE
How they're built — with names attached.
No black boxes. Every agent ships with the corpus it was trained on, the model family it runs on, and the evaluation method that catches drift before you do.
| Agent | Training corpus | Model family | Evaluation method | Last audit |
|---|---|---|---|---|
LOT 042 Anders Sales | 1.2M B2B emails (consented), 38k call transcripts | Claude Sonnet · BYO endpoint optional | Weekly held-out reply-rate / blocklist regression | 2026-04-30 |
LOT 047 Hatfield Support | 120k support tickets, 9k macros, 4k escalations | Claude Haiku · GPT 4-class fallback | Daily CSAT, false-resolution sampling 1% | 2026-04-30 |
LOT 051 Vance Research | Curated public sources + 2k buyer briefs | Claude Sonnet + retrieval (hosted) | Citation-resolves rate, peer audit weekly | 2026-04-22 |
LOT 054 Brunel Ops | 1y of internal-tool change events (5 anchor accounts) | Claude Haiku · long-context | Anomaly precision @ 7-day, false-positive trace | 2026-05-01 |
LOT 058 Lermontov Sales | Inbound chat + form transcripts, 60k qualified | Claude Sonnet · streaming | Booking conversion vs. baseline form, weekly | 2026-04-30 |
LOT 061 Whitfield Research | Public competitor docs, pricing pages, hiring boards | Claude Sonnet + retrieval | Read-through rate + cited-claim audit | 2026-05-02 |
Named operators
Each agent is owned by a real human. Their handle is on the page. Drift is their problem first, yours second.
Public eval
A monthly eval log per agent — replies sampled, false-positives traced, audit dates published.
BYO model
Pro and Enterprise can route Anders, Hatfield, or Vance through your VPC endpoint. Trial uses our hosted stack.
LOT 04 · OUTCOMES
Last 30 days. Real numbers. Named operators.
LOT 042 · 30D
Anders shipped 312 first-touch emails for an 8-person fintech.
42% reply rate · 8 booked meetings · zero list-burn flags · operator sign-off Mira Rao.
LOT 047 · 30D
Hatfield closed 71% of inbound tickets without human handoff.
428 tickets / week · CSAT 4.7 · operator sign-off Sara Okereke.
LOT 051 · 30D
Vance shipped a 42-source competitor brief in 28 hours.
9% rework rate · cited claims audited weekly · operator sign-off Theo Kapoor.
LOT 05 · PRICING
A price on the page.
Subscription is the headline. Outcome pricing — per meeting, per resolved ticket — is available on Pro and Enterprise, capped at 1.5×. Crypto stablecoin (USDT/USDC) via NOWPayments is the v1 rail.
TRIAL
Trial
$99
/ month — cancel anytime
- One agent from the catalog
- Up to 500 actions / month
- Single seat
- Weekly digest
Outcome pricing available on Pro and Enterprise; capped at 1.5×.
PRO
Pro
$499
/ month — 6-month commit
- Up to three agents
- 5k actions / month
- Three seats
- Monthly retraining slot
- Named human owner
Outcome pricing available on Pro and Enterprise; capped at 1.5×.
ENTERPRISE
Enterprise
$4,800+
/ year — concierge invoice
- Unlimited catalog access
- 50k actions / month
- Dedicated channel
- Custom retraining
- 99.5% reachability SLO
Outcome pricing available on Pro and Enterprise; capped at 1.5×.
LOT 06 · FAQ
Eight honest answers.
Why crypto checkout in v1?
Stablecoin rails via NOWPayments are faster than card pilots for our buyer cohort — most of them already pay an OpenAI or Anthropic invoice in USD-equivalent. Wire and ACH are coming later this year. We don't take cards in v1.
Can I demo the agent against my own data?
Yes. The Concierge will ask whether you want to demo against your data or ours. Yours is slower (we connect to a sandbox account first); ours is instant.
What is the SLO?
Pro: 99.5% reachability across the agent's surface. Enterprise: hard targets per agent (e.g., SDR — 100 actions/day, 30-day reply rate ≥ 25%). Trial has no SLO.
How do you handle drift and retraining?
Pro includes one monthly retraining slot per agent, owned by the named human operator. Each agent has a public monthly eval log. Enterprise can pin a model version and lock the corpus.
Can I run the model in our VPC?
Pro and Enterprise support BYO endpoint — Bedrock, Vertex, or Azure OpenAI. Trial uses our hosted stack only.
Is this a marketplace?
No. It's a vetted shelf. Six lots in v1, all owned by named operators inside Prin7r. We add lots quarterly; we sunset lots that don't ship.
What's the rev-share for agencies?
Alex's program: 30% rev-share to the agency on Pro and Enterprise referrals, with a co-branded partner page. Email concierge to register.
What if my Trial doesn't ship?
Cancel anytime in the first 14 days, full refund of the Trial fee. After that, Trial is month-to-month — same cancel-anytime policy.