Skip to content

creatop-gtm/apollo-operator / plugins/apollo-operator/skills/business-brief

business-brief

Verified

Build one readable brief.md capturing the business: identity, offer, who buys and why, objections, proof, and voice. The narrative context layer every other skill reads. Use before ICP work, when copy comes out generic, or when the business changes.

FileKindSize
SKILL.mdskill md12.5 KB
connector-only.mdmarkdown10.5 KB
references/brief-template.mdmarkdown3.4 KB

references/brief-template.md

View on GitHub
# brief.md template

The structure for the single-file business brief that `business-brief` produces. Copy the skeleton, fill what you know, mark the rest `[TBD]`. Keep the whole file readable in one sitting: one to two pages is the target, specific beats complete.

Sections exist to be read by an agent mid-task ("what proof do I cite in step 2?" jumps straight to Proof). Keep the headings exactly as below so downstream skills can rely on them.

---

```markdown
# [Business name]: brief

> Last updated: [date]. Read this before touching the account.

## Identity

One paragraph: what the business does, for whom, and the outcome it delivers.
Written like you would explain it to a smart friend, not like the homepage.

- Website: [url]
- One-liner: [what we sell, one sentence]

## Offer

- What is sold: [the product or service, concretely]
- Pricing model: [flat / tiered / usage / custom, rough range if shareable]
- The outbound offer: [the specific CTA or lead magnet cold prospects are
  asked to respond to. This is what sequences point at.]

## Who buys

One short paragraph describing the ideal customer as a story, not a filter
set (the filters live in profile.yaml).

The buying committee:
- **Champion:** [title(s), what they feel day to day]
- **Economic buyer:** [title(s), what they sign off on and why]
- **End user:** [title(s), if different]

## Why they buy

- Pains: [the 2 to 4 problems, in the buyer's words]
- Triggers: [what makes this urgent now: funding, hiring, a tool change,
  a season, a regulation]
- Desired outcome: [what "solved" looks like to them]

## Why they don't

The objections actually heard, each with the honest answer:

- "[objection]" → [the real answer, not the deflection]
- "[objection]" → [answer]

## Proof

Only what is real. Numbers beat adjectives.

- [Case study or result, with the number and the customer type]
- [Named customers or logos that can be referenced]
- [Testimonial lines worth quoting, with attribution]

## Language

- Buyers say: [the words buyers use for the problem and the product]
- We say: [terms the business prefers]
- Never say: [banned words, competitor framings, claims to avoid]
- Tone: [casual / peer-to-peer / formal, one line on voice]

## Competitive frame

- Alternatives buyers consider: [including "do nothing" and spreadsheets]
- Why us, in one line: [the honest differentiator]

## Constraints

Anything outreach must respect: compliance rules, geographies to avoid,
claims that are off-limits, existing customers or partners not to contact.

## Changelog

- [date]: [what changed and why]
```

---

## Filling notes

- **Identity** is the section agents read when everything else is skipped. Make it carry the whole picture.
- **The outbound offer** is not the product. "A free teardown of your current sequence" is an offer; "our platform" is not. If the business has no outbound offer yet, mark it `[TBD]` and flag it: sequences cannot be written without one.
- **Why they don't** is usually empty on a first draft because nobody volunteers objections. Ask directly (interview question in step 3 of the skill). This section writes the best step 2 and step 3 emails.
- **Proof** feeds `apollo-sequence-builder` step 2 (context and credibility). No proof in the brief means no credibility in the sequence, so chase at least one real number.
- **Language** feeds the `voice:` block of `profile.yaml` and the reviewer's robot-copy check in `sequence-reviewer`.
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 472kUpdated Claude CodeCodexUnverifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 322kUpdated Claude CodeCodexVerifiedInstall