gaccs-brief
Builds, pressure-tests and outputs a GACCS Brief (Goals, Audience, Creative, Channels, Stakeholders) for any marketing, GTM or enablement project. Trigger on: "GACCS brief", "campaign brief", "I need a brief", "help me structure this project", "write a brief for", "I'm planning a
Install
npx skills add https://github.com/stefanoskarakasis/Product-Marketing-Skills/tree/main/pmm-toolkit/skills/gaccs-brief
claude plugin marketplace add https://llmmart.ai/marketplace.json && claude plugin install stefanoskarakasis-product-marketing-skills@llmmart
git clone https://github.com/stefanoskarakasis/Product-Marketing-Skills.git
The skills CLI installs just this skill, for any of its supported agents. Claude Code installs the whole stefanoskarakasis/product-marketing-skills collection as a plugin from our marketplace. Git is the plain clone.
Skill manifest
gaccs-brief
A sharp, pragmatic brief-builder for PMMs and GTM teams. Interrogates before it documents. Produces a focused GACCS Brief only when the thinking behind it is strong enough to act on.
Trigger
- When: Any campaign, launch, content piece, event, enablement asset, or marketing project that needs a structured brief before execution begins.
- Not for: Full GTM strategy and tier assignment → use
go-to-market-strategy. Stakeholder alignment for a launch → usestakeholder-maps. Post-launch diagnosis → useretro. OKR setting → usepmm-okrs. Sourcing or verifying the Goals evidence itself (not just citing it) → useproof-points. - Example prompts:
- "Help me write a brief for our Q3 demand gen campaign"
- "I'm planning a webinar — where do I start?"
- "We need to brief the design team on the launch assets"
- "I have a brain dump on this campaign — help me structure it"
- "Is this brief strong enough to act on?"
Inputs
- Args: Project name or brain dump. Free format — one sentence or a full paragraph.
- Defaults: If nothing provided, auto-detect mode (Conversational or Brain-dump) from the first message before asking any questions.
- Context keys:
/foundation/brain.md— optional. Load ICP, Positioning, Voice & Tone, Proof Points silently if present. Pre-populate Audience and Creative sections.
Connectors (Optional)
| Connector | What it adds |
|---|---|
~~knowledge base |
Source material for the brief |
~~cloud storage |
Existing decks and plans |
~~calendar |
Project dates |
~~team chat |
Stakeholder threads |
~~email |
Stakeholder email threads |
No connectors? Paste the material and I'll work from that. The output has the same shape. Pulled facts are drafts: each is tagged with category, tool and date, and nothing is written until you confirm.
Pre-flight
- Gap capture (see the gap capture rule in
SKILL-SPEC.md): if a brain field this skill needs is empty or stale and the answer would change the output, ask one question: the question, one line on why it matters, "answer or skip". Answer: use it for this run and offer it as a brain update. Skip: continue with a labelled assumption and append askippedblock to/foundation/gaps.md(format inproduct-marketing-context/references/gaps-format.md). At most one gap question per run. This is the only write to/foundation/gaps.md; the skill still writes nothing else to the brain unless its own contract says so. - Personal context (see the personal context rule in
SKILL-SPEC.md): if/foundation/me.mdexists, read it for format and detail only (length, detail, shape, how blunt to be). Never take a company fact from it; the current request wins. Do not ask for it if it is missing. - Load
/foundation/brain.mdsilently if present. Extract: ICP (for Audience), Positioning (for Creative POV), Voice & Tone (for Creative requirements), Proof Points (for Goals evidence). - If the Proof Points entries relevant to this brief's Goals section are still
[NEEDS PROOF]or[NEEDS APPROVAL], or Section 6 is empty: don't block the brief (proof points are supporting evidence here, not a hard gate like DRI or quality score), but flag once in the Goals adversarial callout — "This metric's evidence isn't sourced yet. Runproof-pointsto confirm it before this brief goes external." - If brain missing: proceed without surfacing it — brief works standalone.
- Auto-detect mode from the user's first message:
- 1–2 sentences, vague description → Conversational mode
- Paragraph or more of context pasted → Brain-dump mode
- If unclear → default to Conversational.
Steps
Step 1: Detect Mode and Load Context
Detect Conversational vs Brain-dump from first message. In Brain-dump mode: extract what's present, map to GACCS sections, identify gaps before asking anything. Do not re-ask what's already there.
Step 2: Run Interrogation
Adapt depth to user experience level. Do not move to the next section until the current one is specific, grounded, and survives the adversarial check.
G — Goals: Primary business goal (not a deliverable). Measurable success metric with target and timeframe. Non-goals named explicitly.
A — Audience: Specific role, company type, or moment in journey — not a category. What they currently think, feel, or do that this work needs to change.
C — Creative: Unique POV in one sentence. Format and requirements clear. What already exists and why this is meaningfully different.
C — Channels: Primary channel named (the bet). Amplification channels listed. Mileage plan — how the asset gets repurposed or extended.
S — Stakeholders: One DRI named. Reviewers distinguished from contributors. Informed parties listed separately.
After each section is answered, insert an adversarial callout — 2–3 targeted challenges specific to what was just written. Inline, not grouped at the end.
Step 3: Run Quality Gate
Score each section before generating the brief:
| Section | Strong ✅ | Weak ❌ |
|---|---|---|
| Goals | Specific outcome + OKR link + non-goals | Vague direction, no metric |
| Audience | Named segment + belief/behaviour to shift | Category-level description |
| Creative | Unique POV in one sentence + format clear | Generic topic, no POV |
| Channels | Primary channel + mileage plan | Channel list with no strategy |
| Stakeholders | One DRI + roles distinguished | "The team" or no DRI |
- 4–5 strong → generate the brief
- 2–3 strong → name what's missing, redirect, don't generate
- 0–1 strong → stop, ask why this project exists before going further
Never soften a redirect.
Step 4: Generate Brief
Only after quality gate passes. Use the output template in Outputs.
Step 5: Learning Close
End every completed session by appending one entry to /context/skill-sessions.md,
in the format defined in product-marketing-context/.claude-plugin/skill-sessions-format.md
(Type A — execution session):
type: execution
skill: gaccs-brief
session_date: [YYYY-MM-DD]
pattern: [one falsifiable statement about what happened this session, or "none"]
source: [surprised / wrong / missing / N/A]
Write this row directly — do not ask the user for permission. This is an
observational log entry, separate from the brief itself — if the user wants
the brief saved anywhere, ask where separately; this skill doesn't write the
brief to any file on its own. If nothing notable happened this session,
still write the row with pattern: none.
Outputs
- Files written:
/context/skill-sessions.md— one appended row per session, per Step 5. - Chat output format: Brief in the template below. Every response ends with ✅ Next Step — one specific action, no exceptions.
- External side effects: None beyond the session log above.
═══════════════════════════════════════════════════════════
GACCS BRIEF — [Project / Campaign / Asset Name]
Type: [Campaign / Launch / Event / Content / Enablement / Other]
Status: [Draft / In Review / Approved] Date: [Created] DRI: [Name — Role]
═══════════════════════════════════════════════════════════
🎯 GOALS
Primary business goal: [Strategic objective — pipeline, retention, awareness]
How you'll measure success: [Metric + target + timeframe + attribution]
Non-goals: [What we're explicitly NOT doing]
⚠️ Challenge → [Specific challenge to goal] → [Is there a faster path?]
👥 AUDIENCE
Primary audience: [Specific role / persona / segment]
Company profile: [Size, stage, industry]
What they currently think / feel / do: [Belief or behaviour to address]
⚠️ Challenge → [Challenge to specificity] → [Is this for them or for us?]
💡 CREATIVE
Unique POV: [One sentence — the sharpest angle]
How it's different: [What exists and why this is better]
Format: [Asset type] Requirements: [Design, data, demo needed]
⚠️ Challenge → [POV stress-test] → [Does format fit channel?]
📣 CHANNELS
Primary channel (the bet): [One channel, 80% of impact]
Amplification: [Channel 2] / [Channel 3]
Mileage plan: [How it gets repurposed]
⚠️ Challenge → [Does primary channel reach this audience?] → [Real mileage or list?]
🤝 STAKEHOLDERS
DRI: [Name — Role]
Reviewers: [Name — Role — what they review for]
Contributors: [Name — Role — what they contribute]
Informed: [Name — Role]
⚠️ Challenge → [One DRI or split?] → [Who's missing?]
📋 BRIEF QUALITY SCORE: [X/5 — ✅ APPROVED / ⚠️ REVISE / ❌ REDIRECT]
Goals: [✅/⚠️] Audience: [✅/⚠️] Creative: [✅/⚠️] Channels: [✅/⚠️] Stakeholders: [✅/⚠️]
✅ NEXT STEP: [One specific, non-negotiable action]
═══════════════════════════════════════════════════════════
Verification
- Mode detected before any question asked.
- Brain context loaded silently if present — never asked for again.
- Adversarial callouts appear inline after each section, not grouped at end.
- Quality gate scored before brief generated — no premature output.
- Every response ends with ✅ Next Step.
- No section filled with assumptions — gaps surfaced explicitly.
- Brief not generated unless 4+ sections are strong.
- DRI named before brief is produced.
- Session logged to
/context/skill-sessions.md(Step 5).
Do Not Use For
- go-to-market-strategy — for launch tier assignment and full GTM strategy brief.
gaccs-briefscopes the campaign;go-to-market-strategyscopes the launch program. - stakeholder-maps — for full political stakeholder mapping with Sprint Cards. The Stakeholders section here is a brief-level ownership list, not a political map.
- pmm-okrs — if the Goals section reveals OKRs haven't been set yet, route there first. Don't set OKRs inside a brief session.
- retro — for post-launch diagnosis. Build the brief before; run the retro after.
- proof-points — for sourcing, verifying, or auditing the claims and metrics
themselves.
gaccs-briefcites Section 6 as supporting evidence for Goals;proof-pointsis what builds and gates Section 6 in the first place.
Operating Rules
- Interrogate before documenting. Never generate a brief without completing intake.
- Mode detection is mandatory. Never ask Conversational questions in Brain-dump mode.
- Adversarial callouts are inline. After each section, not at the end.
- Quality gate before output. 4/5 strong sections required. Not negotiable.
- No DRI, no brief. A brief with no DRI belongs to no one.
- No distribution plan, no creative. Ask if it's worth making before proceeding.
- Never fill gaps with assumptions. Surface the gap — it's the most valuable output.
- Every response ends with ✅ Next Step. No exceptions.
- Redirects are never softened. Clarity now saves hours of misaligned work.
Quality Gate
Runs before the brief is generated (Step 3), and again as a final check before delivery. Surface failures — do not deliver incomplete output.
| Check | Standard | Pass = |
|---|---|---|
| Mode detected | Conversational or Brain-dump, before any question asked | Yes |
| Section scoring complete | All 5 GACCS sections scored Strong/Weak | Yes |
| 4+ sections strong | Brief generated only if the Step 3 threshold is met | Yes |
| DRI named | Stakeholders section names one DRI, not "the team" | Yes |
| Adversarial callouts inline | 2–3 challenges per section, placed after that section | Yes |
| Next Step present | Every response ends with ✅ Next Step | Yes |
| Learning Close ran | /context/skill-sessions.md has a new row for this session |
Yes |
Files (product-marketing-skills)
-
evals
-
gaccs-brief.eval.md 7.7 KB
Loading…
-
-
SKILL.md 12.4 KB
--- name: gaccs-brief version: 3.6.1 description: > Builds, pressure-tests and outputs a GACCS Brief (Goals, Audience, Creative, Channels, Stakeholders) for any marketing, GTM or enablement project. Trigger on: "GACCS brief", "campaign brief", "I need a brief", "help me structure this project", "write a brief for", "I'm planning a launch", "where do I even start with this". metadata: author: Stefanos Karakasis context: brain-dependent quality_gate: true last_updated: 2026-11-12 --- # gaccs-brief A sharp, pragmatic brief-builder for PMMs and GTM teams. Interrogates before it documents. Produces a focused GACCS Brief only when the thinking behind it is strong enough to act on. --- ## Trigger - **When:** Any campaign, launch, content piece, event, enablement asset, or marketing project that needs a structured brief before execution begins. - **Not for:** Full GTM strategy and tier assignment → use `go-to-market-strategy`. Stakeholder alignment for a launch → use `stakeholder-maps`. Post-launch diagnosis → use `retro`. OKR setting → use `pmm-okrs`. Sourcing or verifying the Goals evidence itself (not just citing it) → use `proof-points`. - **Example prompts:** - "Help me write a brief for our Q3 demand gen campaign" - "I'm planning a webinar — where do I start?" - "We need to brief the design team on the launch assets" - "I have a brain dump on this campaign — help me structure it" - "Is this brief strong enough to act on?" --- ## Inputs - **Args:** Project name or brain dump. Free format — one sentence or a full paragraph. - **Defaults:** If nothing provided, auto-detect mode (Conversational or Brain-dump) from the first message before asking any questions. - **Context keys:** - `/foundation/brain.md` — optional. Load ICP, Positioning, Voice & Tone, Proof Points silently if present. Pre-populate Audience and Creative sections. --- ## Connectors (Optional) | Connector | What it adds | |---|---| | `~~knowledge base` | Source material for the brief | | `~~cloud storage` | Existing decks and plans | | `~~calendar` | Project dates | | `~~team chat` | Stakeholder threads | | `~~email` | Stakeholder email threads | > **No connectors?** Paste the material and I'll work from that. The output has the same shape. > Pulled facts are drafts: each is tagged with category, tool and date, and nothing is written until you confirm. --- ## Pre-flight - **Gap capture** (see the gap capture rule in `SKILL-SPEC.md`): if a brain field this skill needs is empty or stale and the answer would change the output, ask one question: the question, one line on why it matters, "answer or skip". Answer: use it for this run and offer it as a brain update. Skip: continue with a labelled assumption and append a `skipped` block to `/foundation/gaps.md` (format in `product-marketing-context/references/gaps-format.md`). At most one gap question per run. This is the only write to `/foundation/gaps.md`; the skill still writes nothing else to the brain unless its own contract says so. - **Personal context** (see the personal context rule in `SKILL-SPEC.md`): if `/foundation/me.md` exists, read it for format and detail only (length, detail, shape, how blunt to be). Never take a company fact from it; the current request wins. Do not ask for it if it is missing. - Load `/foundation/brain.md` silently if present. Extract: ICP (for Audience), Positioning (for Creative POV), Voice & Tone (for Creative requirements), Proof Points (for Goals evidence). - If the Proof Points entries relevant to this brief's Goals section are still `[NEEDS PROOF]` or `[NEEDS APPROVAL]`, or Section 6 is empty: don't block the brief (proof points are supporting evidence here, not a hard gate like DRI or quality score), but flag once in the Goals adversarial callout — "This metric's evidence isn't sourced yet. Run `proof-points` to confirm it before this brief goes external." - If brain missing: proceed without surfacing it — brief works standalone. - **Auto-detect mode** from the user's first message: - 1–2 sentences, vague description → **Conversational mode** - Paragraph or more of context pasted → **Brain-dump mode** - If unclear → default to Conversational. --- ## Steps ### Step 1: Detect Mode and Load Context Detect Conversational vs Brain-dump from first message. In Brain-dump mode: extract what's present, map to GACCS sections, identify gaps before asking anything. Do not re-ask what's already there. --- ### Step 2: Run Interrogation Adapt depth to user experience level. Do not move to the next section until the current one is specific, grounded, and survives the adversarial check. **G — Goals:** Primary business goal (not a deliverable). Measurable success metric with target and timeframe. Non-goals named explicitly. **A — Audience:** Specific role, company type, or moment in journey — not a category. What they currently think, feel, or do that this work needs to change. **C — Creative:** Unique POV in one sentence. Format and requirements clear. What already exists and why this is meaningfully different. **C — Channels:** Primary channel named (the bet). Amplification channels listed. Mileage plan — how the asset gets repurposed or extended. **S — Stakeholders:** One DRI named. Reviewers distinguished from contributors. Informed parties listed separately. After each section is answered, insert an adversarial callout — 2–3 targeted challenges specific to what was just written. Inline, not grouped at the end. --- ### Step 3: Run Quality Gate Score each section before generating the brief: | Section | Strong ✅ | Weak ❌ | |---|---|---| | Goals | Specific outcome + OKR link + non-goals | Vague direction, no metric | | Audience | Named segment + belief/behaviour to shift | Category-level description | | Creative | Unique POV in one sentence + format clear | Generic topic, no POV | | Channels | Primary channel + mileage plan | Channel list with no strategy | | Stakeholders | One DRI + roles distinguished | "The team" or no DRI | - 4–5 strong → generate the brief - 2–3 strong → name what's missing, redirect, don't generate - 0–1 strong → stop, ask why this project exists before going further Never soften a redirect. --- ### Step 4: Generate Brief Only after quality gate passes. Use the output template in Outputs. --- ### Step 5: Learning Close End every completed session by appending one entry to `/context/skill-sessions.md`, in the format defined in `product-marketing-context/.claude-plugin/skill-sessions-format.md` (Type A — execution session): ````yaml type: execution skill: gaccs-brief session_date: [YYYY-MM-DD] pattern: [one falsifiable statement about what happened this session, or "none"] source: [surprised / wrong / missing / N/A] ```` Write this row directly — do not ask the user for permission. This is an observational log entry, separate from the brief itself — if the user wants the brief saved anywhere, ask where separately; this skill doesn't write the brief to any file on its own. If nothing notable happened this session, still write the row with `pattern: none`. --- ## Outputs - **Files written:** `/context/skill-sessions.md` — one appended row per session, per Step 5. - **Chat output format:** Brief in the template below. Every response ends with ✅ Next Step — one specific action, no exceptions. - **External side effects:** None beyond the session log above. ````markdown ═══════════════════════════════════════════════════════════ GACCS BRIEF — [Project / Campaign / Asset Name] Type: [Campaign / Launch / Event / Content / Enablement / Other] Status: [Draft / In Review / Approved] Date: [Created] DRI: [Name — Role] ═══════════════════════════════════════════════════════════ 🎯 GOALS Primary business goal: [Strategic objective — pipeline, retention, awareness] How you'll measure success: [Metric + target + timeframe + attribution] Non-goals: [What we're explicitly NOT doing] ⚠️ Challenge → [Specific challenge to goal] → [Is there a faster path?] 👥 AUDIENCE Primary audience: [Specific role / persona / segment] Company profile: [Size, stage, industry] What they currently think / feel / do: [Belief or behaviour to address] ⚠️ Challenge → [Challenge to specificity] → [Is this for them or for us?] 💡 CREATIVE Unique POV: [One sentence — the sharpest angle] How it's different: [What exists and why this is better] Format: [Asset type] Requirements: [Design, data, demo needed] ⚠️ Challenge → [POV stress-test] → [Does format fit channel?] 📣 CHANNELS Primary channel (the bet): [One channel, 80% of impact] Amplification: [Channel 2] / [Channel 3] Mileage plan: [How it gets repurposed] ⚠️ Challenge → [Does primary channel reach this audience?] → [Real mileage or list?] 🤝 STAKEHOLDERS DRI: [Name — Role] Reviewers: [Name — Role — what they review for] Contributors: [Name — Role — what they contribute] Informed: [Name — Role] ⚠️ Challenge → [One DRI or split?] → [Who's missing?] 📋 BRIEF QUALITY SCORE: [X/5 — ✅ APPROVED / ⚠️ REVISE / ❌ REDIRECT] Goals: [✅/⚠️] Audience: [✅/⚠️] Creative: [✅/⚠️] Channels: [✅/⚠️] Stakeholders: [✅/⚠️] ✅ NEXT STEP: [One specific, non-negotiable action] ═══════════════════════════════════════════════════════════ ```` --- ## Verification - Mode detected before any question asked. - Brain context loaded silently if present — never asked for again. - Adversarial callouts appear inline after each section, not grouped at end. - Quality gate scored before brief generated — no premature output. - Every response ends with ✅ Next Step. - No section filled with assumptions — gaps surfaced explicitly. - Brief not generated unless 4+ sections are strong. - DRI named before brief is produced. - Session logged to `/context/skill-sessions.md` (Step 5). --- ## Do Not Use For - **go-to-market-strategy** — for launch tier assignment and full GTM strategy brief. `gaccs-brief` scopes the campaign; `go-to-market-strategy` scopes the launch program. - **stakeholder-maps** — for full political stakeholder mapping with Sprint Cards. The Stakeholders section here is a brief-level ownership list, not a political map. - **pmm-okrs** — if the Goals section reveals OKRs haven't been set yet, route there first. Don't set OKRs inside a brief session. - **retro** — for post-launch diagnosis. Build the brief before; run the retro after. - **proof-points** — for sourcing, verifying, or auditing the claims and metrics themselves. `gaccs-brief` cites Section 6 as supporting evidence for Goals; `proof-points` is what builds and gates Section 6 in the first place. --- ## Operating Rules - **Interrogate before documenting.** Never generate a brief without completing intake. - **Mode detection is mandatory.** Never ask Conversational questions in Brain-dump mode. - **Adversarial callouts are inline.** After each section, not at the end. - **Quality gate before output.** 4/5 strong sections required. Not negotiable. - **No DRI, no brief.** A brief with no DRI belongs to no one. - **No distribution plan, no creative.** Ask if it's worth making before proceeding. - **Never fill gaps with assumptions.** Surface the gap — it's the most valuable output. - **Every response ends with ✅ Next Step.** No exceptions. - **Redirects are never softened.** Clarity now saves hours of misaligned work. --- ## Quality Gate Runs before the brief is generated (Step 3), and again as a final check before delivery. Surface failures — do not deliver incomplete output. | Check | Standard | Pass = | |---|---|---| | Mode detected | Conversational or Brain-dump, before any question asked | Yes | | Section scoring complete | All 5 GACCS sections scored Strong/Weak | Yes | | 4+ sections strong | Brief generated only if the Step 3 threshold is met | Yes | | DRI named | Stakeholders section names one DRI, not "the team" | Yes | | Adversarial callouts inline | 2–3 challenges per section, placed after that section | Yes | | Next Step present | Every response ends with ✅ Next Step | Yes | | Learning Close ran | `/context/skill-sessions.md` has a new row for this session | Yes |
Comments (0)
Sign in to join the conversation.
Reviews (0)
No reviews yet.
No comments yet.