v1.0 to v1.0.0

75 added, 78 removed. Audit A to A.

---

name: tov-guidelines

- version: '1.0'

- last_updated: 2026-05-19

- author: genesys-growth

- description: Extract voice patterns from existing content (website, blog, social, sales-call transcripts) and codify into voice rules. Produces voice analysis + voice guidelines (rules per pattern + violation pattern + fix template). Writes to marketing/brand/brand-voice.md as the canonical voice rules every content skill reads. Triggers - "tone of voice", "brand voice", "voice guidelines", "TOV audit", "writing rules", "extract voice"

- goal: Analyze existing content (website, blog, social, sales-call transcripts) and codify voice patterns into 5 voice rules with violation/fix patterns.

- outcome: marketing/brand/brand-voice.md with 5 voice rules every content skill applies.

- primitive: research

- ontology_type: tone-of-voice

- review_gate: 2

- inputs:

- required: []

- recommended:

- []

- outputs:

- - type: tone-of-voice

- feeds_into:

- - content-strategy

- - landing-page-copy

- - product-messaging

- owned_by_agent: researcher

- mcps_used:

- - exa

- triggers:

- slash_commands:

- - /tov-guidelines

- status: draft

+ description: >-

+ Extracts a company's tone of voice from its own content and turns it into rules a writer or an AI can

+ follow. Phase 1 analyses 15 to 20 pages for sentence, paragraph, word and structure patterns with

+ frequency counts; after the user reviews it, Phase 2 writes the guidelines: tone attributes with

+ before and after examples, a We are / We are not table, tone by context, vocabulary to use and avoid,

+ and open questions. Triggers: "tone of voice", "brand voice", "voice guidelines", "writing style

+ guide", "extract voice from our website", "TOV". Not for visual identity (use /brand-kit) or messaging

+ (use /messaging).

+ version: '1.0.0'

+ license: MIT

---

- # tov-guidelines — Stage 1 + 2 research skill (PMM spine)

- Analyzes existing content (your website, blog, recent LinkedIn posts, sales-call transcripts) and produces voice rules that downstream content skills apply.

- ---

+ # TOV guidelines

- ## When to use

+ Find out how a company actually writes, then write that down as rules specific enough that a new writer, or an AI, sounds like them on the first draft. Two phases with a review in between: first the evidence, then the guidelines.

- - When onboarding a new content writer / contractor — they need codified voice rules to write in your voice

- - When LinkedIn or blog output stops sounding like the founder / brand

- - Quarterly refresh

+ ## When to use it

- ## When NOT to use

+ - Building the context folder, before any copy or content work.

+ - Starting a content program, or onboarding writers or AI tools.

+ - When content has drifted, or the existing guide is more than six months old.

- - For visual brand identity (use `/brand-kit` — voice + visual are separate but both live in `marketing/brand/`)

- - For one-off content checks against existing rules (write a custom gate skill that reads `brand-voice.md`)

- - For positioning or messaging (use `/positioning` or `/product-messaging`)

+ Don't use it for colours and fonts (`/brand-kit`), for what to say (`/messaging`), or when the user only wants a one-line description of the voice.

- ## How it works

+ ## Inputs

- 1. Inputs: source content (one or more of: website URL, blog URL, LinkedIn handle, sales-call transcripts, founder writing samples).

- 2. Reads (Exa MCP for fresh data): the source content. Patterns extracted include: vocabulary preferences, sentence-length defaults, headline conventions, em-dash usage, list-vs-prose preference, contraction frequency, persona pronouns, signature opening / closing patterns.

- 3. Produces a two-phase output:

- - **Phase 1 — Voice analysis** (extracted patterns from source material with frequency scores): "uses em-dash with spaces 87% of the time across 30 LinkedIn posts; sentence case headings 100% of the time; banned-word violations 0 in last 10 posts"

- - **Phase 2 — Voice guidelines** (codified rules every content skill applies):

- - Rule N — {rule name}

- - **The rule:** 1-sentence statement

- - **Why:** 1-sentence justification

- - **Violation pattern:** what breaking this looks like

- - **Fix template:** how to fix

- - **Exception:** when the rule doesn't apply

- 4. Writes to `marketing/brand/brand-voice.md` (overwrites prior canonical; git history preserves prior versions). Marks `status: locked` once approved.

+ **Required:** the company website URL. It needs a homepage and at least 5 to 10 pages of real content. If it's thin, say so and flag the sample size in the output.

- ## Invoke

+ **Optional:** founder notes or transcripts, an existing style guide to update, `context/icp/` for the reader, `context/expert-pov/` for the founder's own voice.

- ```

- /tov-guidelines https://yourdomain.com + analyze my last 10 LinkedIn posts

- ```

+ **Paths.** Write `context/brand/MMYY-tov-analysis.md` (Phase 1) and `context/brand/MMYY-tov-guidelines.md` (Phase 2). Agency or consultant? Swap `context/` for `clients/<client>/context/` throughout.

- Or open-ended:

- ```

- /tov-guidelines extract voice from these sales-call transcripts: [paths]

- ```

+ **Tools.** Page fetching for the site. Web search with a site filter helps collect founder posts. Exa or Firecrawl are fine if you have them.

- ## Example output

+ ## Phase 1: analysis

- See [`marketing/brand/brand-voice.md`](../../../marketing/brand/brand-voice.md) for the PulseAnalytics seed. Notice the 5 rules each have rule statement + why + violation pattern + fix template + exception. The 5 rules cluster into two themes (prose hygiene + claim integrity).

+ 1. **Pick the pages.** Homepage, about or values, 3 to 5 blog posts, 2 to 3 case studies, pricing, FAQ, a landing page. Aim for 15 to 20.

+ 2. **Sentence patterns:** average length, first or third person, questions, imperatives.

+ 3. **Paragraph patterns:** length, how they open, where evidence sits.

+ 4. **Word patterns:** the company's own terms, the customer's terms, words they avoid, how often they use modifiers.

+ 5. **Structure patterns:** sentence or title case headers, CTA wording and placement, how they stack proof.

+ 6. **Score frequency.**

+ - High: 80% of pages or more.

+ - Medium: 40 to 79%.

+ - Low: under 40%.

+ - Conflict: pages contradict each other.

- ## Dependencies

+ Show the count ("5 of 6 pages").

+ 7. **Map voice by content type.** Person, formality and CTA style for homepage, blog, case study and pricing.

+ 8. **Collect real examples** for each pattern. Use real sentences only, never invented ones.

+ 9. **Flag inconsistencies and gaps**, with the questions you'd ask the founder.

+ 10. **Review gate.** The user confirms the patterns, corrects anything wrong and answers the gap questions. Don't start Phase 2 without this.

- - **Reads from:** source content (URL or files — provided as input); optionally `marketing/icp/ICP.md` (voice should match the ICP's vocabulary register)

- - **Reads via Exa MCP (optional):** the source content if it lives on the web

- - **Writes to:** `marketing/brand/brand-voice.md` (canonical)

- - **Downstream readers:** any content skill (LinkedIn, blog, newsletter, email), `/landing-page-copy`, future gate skills

+ ## Phase 2: guidelines

- ## Customization

+ 1. **Apply the feedback** and resolve each inconsistency.

+ 2. **Write the core guide:**

+ - who reads it

+ - 3 to 5 tone attributes, each with a before and after rewrite

+ - vocabulary to use and avoid

+ - structure templates for the main content types

+ 3. **We are / We are not table.** 5 to 8 rows, each backed by Phase 1 evidence.

+ 4. **Tone by context.** How the voice shifts for these contexts:

+ - cold outreach

+ - nurture email

+ - objection handling

+ - sales follow-up

+ - support reply

+ - case study

+ - social post

+ 5. **Tone by reader emotion.** How to write to someone frustrated, sceptical, excited or rushed.

+ 6. **AI writing tells to avoid.** At least five patterns that would make this brand sound machine-written, shown with the brand's own fix. Examples: generic praise openers, three short parallel sentences, "it's not X, it's Y" contrasts, filler transitions, closing on "what do you think?".

+ 7. **Optional sections**, only when they apply:

+ - hard questions answered: security, pricing, a big competitor. Mark these draft until the founder approves.

+ - product microcopy, for companies with a product UI.

+ - AI assistant voice, for companies with a chatbot.

+ 8. **Open questions.** At least three, each saying what it unblocks.

+ 9. **Source every rule** to a page, or mark it "TBD, needs founder input".

- The 5-rule default in the PulseAnalytics example is operator-flavored. Yours may need more rules (large enterprise teams sometimes track 8-12) or fewer (founder voice often clusters into 3-4 rules). Use whatever rule count matches the patterns you can actually enforce.

+ ## What good looks like

- Don't add a rule you can't enforce — every rule should map to a check a gate skill could run automatically. Vague rules ("be human") get ignored; specific rules ("no em-dashes without spaces") get enforced.

+ - At least 15 pages read, with findings in every pattern category.

+ - Frequencies are counts, never "usually" or "often".

+ - No adjective-only rules. "Friendly" isn't a guideline; "use 'you' in the first sentence of every page" is.

+ - Every example is real text from the site.

+ - A writer could pick the right tone for a support reply without asking anyone.

- ## Where this fits in the Example 1 chain

+ ## Mistakes to avoid

- ```

- Week 1: /tov-guidelines (parallel with competitor + ICP research) → voice rules locked

- Week 2: /positioning + /product-messaging apply voice rules to output

- Week 3+: every content / paid / outbound execution skill reads voice rules

- ```

+ - Writing the voice you wish they had instead of the one they have. Note the gap; don't paper over it.

+ - Skipping the review gate.

+ - Copying another company's style guide structure and filling it with adjectives.

+ - Rules so vague any brand could use them.

- ## Refresh cadence

+ ## Chain

- Quarterly. Trigger sooner when the team finds a new buzzword pattern, customer flags off-brand phrasing, or a new persona is added (vocabulary may shift).

+ Step 6 of the suggested chain: `/company-context` → `/icp-research` → `/competitor-research` → `/win-loss` → `/expert-pov` → **tov-guidelines** → `/brand-kit` → `/positioning` → `/messaging` → `/funnel-strategy` → `/goals`.