Prospect Desk
Find a small number of suitable prospects and prepare outreach that a person can send.
Set up the crew and pass the results between Bots yourself. The recorded sample uses simulated text handoffs; separate Bots have not run it in Hermes or Grok.
When to use this crew
Use this crew for repeated outreach: research prospects, check their fit, draft messages and avoid contacting the same person twice. For one warm introduction, one Bot with the relevant background may be enough.
The jobs
- Researcher
Verify the approved accounts and offer-relevant facts.
- Lead Qualifier
Apply fit rules and contact restrictions.
- Writer
Draft short messages from verified facts.
- Outreach Coordinator
Check duplicates and stage a limited, unsent batch.
What you will need
Offer, target criteria, approved account list and relationship notes, current conversations, suppression list, contact log, batch limit, and sender.
Set up and run
- Select only the roles this job needs. Existing role links lead to Cabinet's current setup pages; new role links contain full manual-test instructions.
- Put each selected role in its own conversation or Bot. Paste the new role's Role instructions section, or add the task extension below to an existing role. These files are not native import archives or add-to-app links.
- Start with the fictional first task below. Supply the same approved source records to each role that needs them. A copied handoff alone is not evidence for a factual claim.
- Pass reviewed results in the order below. If your app lacks Bot-to-Bot handoff, copy the small handoff yourself. Do not assume a shared folder, tool connection, or group conversation creates separate security boundaries.
- Record each actual answer and failed attempt. Review before connecting real accounts or adding a schedule. Set read-only access and service-side cost caps where available; these written limits cannot enforce a billing cap.
Pass the work in this order
- Researcher provides verified account records and unknowns.
- Lead Qualifier labels fit, hold, or exclude, retaining reasons.
- Writer drafts only for approved fit accounts.
- Outreach Coordinator checks the newest contact log and presents unsent drafts to the person.
Use this handoff
- Task and current version:
- Approved source records:
- Result and its status: draft, checked, or approved:
- Unresolved questions:
- Next owner and action:
- External action awaiting specific approval:
If a role cannot finish, it returns the missing input or failed attempt. The next role must not manufacture a success record.
First task
Fictional target: tutoring firms with ten or more tutors. A has twelve but opted out. B has six. C is Cedar Tutors, with eighteen tutors and no prior contact; its supplied contact is Riley. D has fifteen but is in an active sales conversation. Approved offer and sole product claim: software that sends appointment reminders. Sender: Avery. Channel: email draft in this conversation, with no email address supplied and no external sending. Batch cap: two. Draft only for eligible accounts.
Check the result
- Excludes A and B and holds D.
- Returns only C, even though the limit permits two.
- Invents no contact address, relationship, or intent.
- All messages remain unsent and no live sequences change.
Read the sample answer and checks
Sample written in Codex on October 6, 2026. The same assistant wrote and checked the answer. This is not an independent review or a Hermes or Grok app test.
Original answer
Simulated text-only crew result. No separate Bot conversations or native crew run occurred.
Eligible batch: one draft for C, Cedar Tutors, within the cap of two. A is excluded because it opted out, despite having twelve tutors. B is excluded because six tutors fall below ten. D is held because its active sales conversation conflicts with new outreach. C meets the size rule with eighteen tutors and the supplied record states no prior contact; this establishes fit, not buying intent.
Email draft — sender Avery; recipient Riley at Cedar Tutors; email address not supplied.
Subject: Appointment reminders for Cedar Tutors Hi Riley, Our software sends appointment reminders for tutoring firms. Would you like to discuss whether it fits Cedar Tutors? Avery
Status: drafted, not approved, not sent. Do not add a second contact to fill the cap. The proposed message uses only the supplied product claim and account/contact names. It makes no claims about no-shows, savings, automation, or Riley's interests.
Sources: fictional target criterion; A–D account records; sole approved reminder claim; sender, channel, and batch cap in the First task. These are supplied snapshots, not live research. Missing: C's email address and approval of the exact message before external sending; any future batch needs the latest contact and suppression records. Next owner: Avery to review the draft. Sending is explicitly outside this task. No accounts were connected or external records changed.
Checklist results
- pass: Excludes A and B and holds D. — Actual answer excludes A for opt-out, B for six tutors, and holds D for active sales conversation.
- pass: Returns only C, even though the limit permits two. — Only Cedar Tutors/Riley receives a draft; the second slot is not filled.
- pass: Invents no contact address, relationship, or intent. — Email address is explicitly missing; no invented relationship or buying interest is asserted.
- pass: All messages remain unsent and no live sequences change. — Draft status is explicitly unapproved/unsent; execution used only reads and local exercise-record writes.
Not tested
- Explicitly simulated text handoffs in one Codex subagent; no separate Bot conversations or native crew runtime tested.
- No operational tools, external account access, paid calls, spending, publishing, or live record mutations exercised.
- Fictional records only; missing real inputs were not fabricated.
Task instructions for existing Bots and run limits
Researcher
Keep the existing Bot's role and boundaries. For this crew: Verify the approved accounts and offer-relevant facts. Check a defined question against the original approved records. Return evidence, source references, and gaps. Label supplied snapshots separately from current live research. Use only the input assigned to you for this task. Return the result, supporting records, unresolved questions, and next owner. A handoff does not approve an external action.
Writer
Keep the existing Bot's role and boundaries. For this crew: Draft short messages from verified facts. Draft for the named audience using the approved brief. Preserve meaning and claims; never invent quotations or evidence. Keep drafts unsent. Use only the input assigned to you for this task. Return the result, supporting records, unresolved questions, and next owner. A handoff does not approve an external action.
Run limits
Maximum ten supplied accounts and two draft messages in the sample. No paid enrichment, outbound sending, or CRM changes.
Stop when the task finishes, a required input is missing, a tool fails, or an action needs new approval. No automatic retries. No new credentials, private connectors, external messages, purchases, publishing, or deployment are part of the sample.
Finished output
A qualified shortlist, exclusion reasons, and a small unsent message set.
Workflow written by Bot Cabinet. Inspired by The Outbound Crew, from the supplied @unicodef1wn chart. Read the source notes. See Grok’s official guides for setup and sharing instructions.