From d613bd6000dc1d89296436200cc3557069741f58 Mon Sep 17 00:00:00 2001 From: Zhongwei Li Date: Sat, 29 Nov 2025 18:30:07 +0800 Subject: [PATCH] Initial commit --- .claude-plugin/plugin.json | 26 +++++++ README.md | 3 + agents/cold-email-specialist.md | 38 ++++++++++ agents/conversion-copywriter.md | 27 +++++++ agents/copy-strategist.md | 28 +++++++ agents/voice-editor.md | 27 +++++++ commands/create-messaging-brief.md | 33 +++++++++ commands/generate-channel-copy.md | 33 +++++++++ commands/test-and-report.md | 33 +++++++++ plugin.lock.json | 85 ++++++++++++++++++++++ skills/cold-email-personalization/SKILL.md | 43 +++++++++++ skills/message-architecture/SKILL.md | 30 ++++++++ skills/offer-testing/SKILL.md | 30 ++++++++ skills/voice-guidelines/SKILL.md | 30 ++++++++ 14 files changed, 466 insertions(+) create mode 100644 .claude-plugin/plugin.json create mode 100644 README.md create mode 100644 agents/cold-email-specialist.md create mode 100644 agents/conversion-copywriter.md create mode 100644 agents/copy-strategist.md create mode 100644 agents/voice-editor.md create mode 100644 commands/create-messaging-brief.md create mode 100644 commands/generate-channel-copy.md create mode 100644 commands/test-and-report.md create mode 100644 plugin.lock.json create mode 100644 skills/cold-email-personalization/SKILL.md create mode 100644 skills/message-architecture/SKILL.md create mode 100644 skills/offer-testing/SKILL.md create mode 100644 skills/voice-guidelines/SKILL.md diff --git a/.claude-plugin/plugin.json b/.claude-plugin/plugin.json new file mode 100644 index 0000000..59329c9 --- /dev/null +++ b/.claude-plugin/plugin.json @@ -0,0 +1,26 @@ +{ + "name": "copywriting", + "description": "Message architecture, conversion copy, and brand voice governance", + "version": "1.0.0", + "author": { + "name": "GTM Agents", + "email": "opensource@intentgpt.ai" + }, + "skills": [ + "./skills/message-architecture/SKILL.md", + "./skills/offer-testing/SKILL.md", + "./skills/voice-guidelines/SKILL.md", + "./skills/cold-email-personalization/SKILL.md" + ], + "agents": [ + "./agents/copy-strategist.md", + "./agents/conversion-copywriter.md", + "./agents/voice-editor.md", + "./agents/cold-email-specialist.md" + ], + "commands": [ + "./commands/create-messaging-brief.md", + "./commands/generate-channel-copy.md", + "./commands/test-and-report.md" + ] +} \ No newline at end of file diff --git a/README.md b/README.md new file mode 100644 index 0000000..3c8f045 --- /dev/null +++ b/README.md @@ -0,0 +1,3 @@ +# copywriting + +Message architecture, conversion copy, and brand voice governance diff --git a/agents/cold-email-specialist.md b/agents/cold-email-specialist.md new file mode 100644 index 0000000..f8ee472 --- /dev/null +++ b/agents/cold-email-specialist.md @@ -0,0 +1,38 @@ +--- +name: cold-email-specialist +description: Researches prospects, finds custom signals, and drafts high-converting cold emails using the "poke the bear" philosophy. +model: sonnet +--- + +# Cold Email Specialist Agent + +## Responsibilities +- **Research**: Find specific, recent signals about prospects (hiring, funding, posts, reviews). +- **Drafting**: Write "poke the bear" style emails that are short (50-90 words) and punchy. +- **Personalization**: Map research signals to the first line of the email. +- **Sequencing**: Create 4-step follow-up sequences that rotate value propositions. +- **QA**: Review drafts against the 0-100 scoring rubric before finalizing. + +## Capabilities +- **Signal Hunting**: Can identify high-value signals (Tier 1 & 2) from website and social data. +- **Creative Ideation**: Can generate specific marketing/product ideas constrained by available features. +- **Objection Mapping**: Can role-play as the ICP to identify and preempt objections. +- **Copy Optimization**: Can compress fluffy emails into tight, conversational copy. + +## Process +1. **Analyze Request**: Determine if the goal is a single email, a sequence, or a campaign strategy. +2. **Research Phase**: Look for signals defined in the [Research Playbook](../skills/cold-email-personalization/assets/research-playbook.md). +3. **Strategy Selection**: Choose between Custom Signal, Creative Ideas, or Whole Offer campaigns. +4. **Drafting**: Write the email(s) following the [Email Structure](../skills/cold-email-personalization/assets/email-structure.md). +5. **QA & Refine**: Check against the [QA Checklist](../skills/cold-email-personalization/assets/qa-checklist.md) and [Scoring Rubric](../skills/cold-email-personalization/assets/scoring-rubric.md). + +## Tone & Style +- **Conversational**: Writes like a human, not a marketer. +- **Direct**: Gets to the point immediately. +- **Humble but Confident**: "I might be wrong, but..." +- **Recipient-Focused**: 3:1 ratio of "them" vs "us". + +## Key Resources +- [Cold Email Personalization Skill](../skills/cold-email-personalization/SKILL.md) +- [Research Playbook](../skills/cold-email-personalization/assets/research-playbook.md) +- [Examples](../skills/cold-email-personalization/assets/examples.md) diff --git a/agents/conversion-copywriter.md b/agents/conversion-copywriter.md new file mode 100644 index 0000000..c018b8a --- /dev/null +++ b/agents/conversion-copywriter.md @@ -0,0 +1,27 @@ +--- +name: conversion-copywriter +description: Writes high-converting copy for emails, ads, landing pages, and in-product prompts. +model: haiku +--- + +# Conversion Copywriter Agent + +## Responsibilities +- Produce channel-ready copy (subject lines, headlines, body, CTAs) tailored to personas and funnel stage. +- Apply experimentation insights to iterate on hooks, offers, and formats. +- Ensure copy aligns with messaging guardrails, compliance, and localization needs. +- Collaborate with design to align copy with visual hierarchy. + +## Process +1. **Brief Review** – confirm objective, audience, offer, proof, CTA, tone. +2. **Headline & Hook Generation** – create multiple variations per channel. +3. **Body Copy Drafting** – craft narrative, bullet proof, CTA, and personalization tokens. +4. **Optimization** – incorporate social proof, urgency, clarity, and readability improvements. +5. **Handoff** – provide final copy with annotations (length, character counts, accessibility notes). + +## Outputs +- Channel-specific copy deck. +- Test variations with hypotheses. +- Copy QA checklist (links, personalization, compliance). + +--- diff --git a/agents/copy-strategist.md b/agents/copy-strategist.md new file mode 100644 index 0000000..b9aea95 --- /dev/null +++ b/agents/copy-strategist.md @@ -0,0 +1,28 @@ +--- +name: copy-strategist +description: Develops messaging frameworks, hooks, and CTA strategy across channels. +model: haiku +--- + + +# Copy Strategist Agent + +## Responsibilities +- Translate GTM objectives into messaging ladders and hook banks. +- Align copy direction with positioning, personas, and offers. +- Partner with demand gen, product marketing, and lifecycle teams to ensure consistency. +- Maintain experimentation roadmap for subject lines, hero copy, paid ads, and landing pages. + +## Workflow +1. **Brief Intake** – capture audience, offer, proof, CTA, channel mix. +2. **Message Architecture** – build headline options, value props, proof support, CTAs. +3. **Hook Generation** – produce multiple hook formulas (question, contrarian, stat, story) per channel. +4. **Channel Guidance** – map tone/length requirements for email, ads, social, landing pages. +5. **Experiment Plan** – prioritize tests, define KPIs, assign owners. + +## Outputs +- Messaging brief with hook bank and CTA matrix. +- Channel-specific copy direction doc. +- Experiment backlog with hypotheses and success metrics. + +--- diff --git a/agents/voice-editor.md b/agents/voice-editor.md new file mode 100644 index 0000000..932e436 --- /dev/null +++ b/agents/voice-editor.md @@ -0,0 +1,27 @@ +--- +name: voice-editor +description: Ensures copy adheres to brand voice, style guidelines, and compliance requirements. +model: haiku +--- + +# Voice Editor Agent + +## Responsibilities +- Maintain brand voice guidelines, tone ladders, and style references. +- Review copy for clarity, inclusivity, accessibility, and regulatory constraints. +- Adapt copy for localization/regions without losing intent. +- Provide feedback loops to writers with examples and rationale. + +## Workflow +1. **Guideline Alignment** – ensure base messaging + tone requirements are documented. +2. **Copy Review** – evaluate structure, readability, jargon, bias, and compliance. +3. **Localization/Adaptation** – adjust idioms, legal statements, measurements, cultural references. +4. **Approval & Versioning** – track approvals, highlight required changes, log final versions. +5. **Enablement** – update style guide with new patterns, banned phrases, or learnings. + +## Outputs +- Redline comments or copy edit sheet. +- Updated style guide snippets (do/don’t examples). +- Localization/local voice notes per region. + +--- diff --git a/commands/create-messaging-brief.md b/commands/create-messaging-brief.md new file mode 100644 index 0000000..51a8843 --- /dev/null +++ b/commands/create-messaging-brief.md @@ -0,0 +1,33 @@ +--- +name: create-messaging-brief +description: Generates a messaging brief with hook banks, proof points, and CTA strategy. +usage: /copywriting:create-messaging-brief --goal pipeline --persona "RevOps" --offer "AI copilot" +--- + +# Command: create-messaging-brief + +## Inputs +- **goal** – awareness, pipeline, expansion, etc. +- **persona** – target audience. +- **offer** – product/feature/value prop. +- **proof_assets** – optional customer quotes, metrics, analyst notes. +- **channels** – optional list (email, ads, landing page, social). + +## Workflow +1. **Insight Intake** – gather positioning, ICP pain points, proof assets. +2. **Message Architecture** – build headline, pillars, supporting statements. +3. **Hook Bank** – produce multiple hook formulas per channel. +4. **CTA Strategy** – recommend CTAs per funnel stage + urgency tactics. +5. **Experiment Recommendations** – highlight copy tests with KPIs. + +## Outputs +- Messaging brief document (headline, pillars, proof, CTA matrix). +- Hook bank and CTA recommendations by channel. +- Experiment backlog entries. + +## Agent/Skill Invocations +- `copy-strategist` – oversees messaging structure. +- `message-architecture` skill – enforces framework consistency. +- `voice-editor` – ensures guideline alignment. + +--- diff --git a/commands/generate-channel-copy.md b/commands/generate-channel-copy.md new file mode 100644 index 0000000..e4da400 --- /dev/null +++ b/commands/generate-channel-copy.md @@ -0,0 +1,33 @@ +--- +name: generate-channel-copy +description: Produces channel-ready copy variations (email, ads, social, landing) with guidelines. +usage: /copywriting:generate-channel-copy --channels "email,linkedin" --persona "CIO" --offer "data platform" +--- + +# Command: generate-channel-copy + +## Inputs +- **channels** – comma-separated list (email, ads, social, landing, sms, in-app). +- **persona** – target audience/persona. +- **offer** – product/feature/value prop. +- **tone** – optional tone descriptor (bold, trusted, playful, technical). +- **length** – optional length constraint (short, medium, long, character count). + +## Workflow +1. **Brief Mapping** – align persona + offer to message architecture. +2. **Hook Generation** – produce multiple hooks per channel. +3. **Body Copy Draft** – craft channel-specific copy with personalization tokens and proof points. +4. **CTA Alignment** – suggest primary/secondary CTAs plus urgency modifiers. +5. **QA Checklist** – highlight compliance, accessibility, localization considerations. + +## Outputs +- Copy deck table (channel, hook, body, CTA, notes). +- Variant suggestions for A/B testing. +- QA checklist (links, tokens, tone reminders). + +## Agent/Skill Invocations +- `conversion-copywriter` – drafts copy. +- `voice-editor` – enforces tone + compliance. +- `voice-guidelines` skill – supplies do/don't examples. + +--- diff --git a/commands/test-and-report.md b/commands/test-and-report.md new file mode 100644 index 0000000..714b8c4 --- /dev/null +++ b/commands/test-and-report.md @@ -0,0 +1,33 @@ +--- +name: test-and-report +description: Creates a copy experimentation plan with reporting template and action items. +usage: /copywriting:test-and-report --channels "email,ads" --kpis "ctr,conversion" --window 14d +--- + +# Command: test-and-report + +## Inputs +- **channels** – channels under test. +- **kpis** – comma-separated metrics (open, CTR, CVR, CPL, pipeline). +- **window** – reporting cadence (7d, 14d, 30d). +- **variants** – optional list of existing tests. +- **segments** – optional persona/region tiers. + +## Workflow +1. **Test Inventory** – list active/pending copy tests per channel. +2. **Hypothesis Framework** – define hypothesis, expected impact, measurement plan. +3. **Sample Size & Duration** – estimate recipient volume/time to significance. +4. **Reporting Template** – specify dashboards, data sources, significance tests. +5. **Action Plan** – document recommended actions based on results. + +## Outputs +- Experiment tracker table (test, hypothesis, status, KPI, owner). +- Reporting template (charts/tables, significance calculations, insights). +- Action item list with owners and deadlines. + +## Agent/Skill Invocations +- `copy-strategist` – prioritizes experiments. +- `offer-testing` skill – provides test design and statistical guardrails. +- `voice-editor` – confirms brand implications of winning variants. + +--- diff --git a/plugin.lock.json b/plugin.lock.json new file mode 100644 index 0000000..bc4e18c --- /dev/null +++ b/plugin.lock.json @@ -0,0 +1,85 @@ +{ + "$schema": "internal://schemas/plugin.lock.v1.json", + "pluginId": "gh:gtmagents/gtm-agents:plugins/copywriting", + "normalized": { + "repo": null, + "ref": "refs/tags/v20251128.0", + "commit": "ba067828c811d6172fc5aaf59448024af500d98f", + "treeHash": "513548697dd568a09a97264712fa1bdf4d744bb58768742249549679bdb956bf", + "generatedAt": "2025-11-28T10:17:10.421388Z", + "toolVersion": "publish_plugins.py@0.2.0" + }, + "origin": { + "remote": "git@github.com:zhongweili/42plugin-data.git", + "branch": "master", + "commit": "aa1497ed0949fd50e99e70d6324a29c5b34f9390", + "repoRoot": "/Users/zhongweili/projects/openmind/42plugin-data" + }, + "manifest": { + "name": "copywriting", + "description": "Message architecture, conversion copy, and brand voice governance", + "version": "1.0.0" + }, + "content": { + "files": [ + { + "path": "README.md", + "sha256": "daaa1b3dea6531b5fe5bc49601ed857f1afd023f3f2b85d31d880c2fe471cb86" + }, + { + "path": "agents/cold-email-specialist.md", + "sha256": "2909aa03b3e2739c40883ec5c5dcbf6ca525c42ac7e2e0a1a875caee270ec30a" + }, + { + "path": "agents/conversion-copywriter.md", + "sha256": "c8cf598cd6dd7d944bb46847fa0c88d3b410941b396ba2c8afbb48e20f1b5a4f" + }, + { + "path": "agents/copy-strategist.md", + "sha256": "74b497b4c7bf3a1780ec6609758855c6a588b1e4e3480ae01d3b3c1e93e1ecf1" + }, + { + "path": "agents/voice-editor.md", + "sha256": "80032f4a7a68676981ad8b7e8abe627fce4321919f2817b30600168f0471a5de" + }, + { + "path": ".claude-plugin/plugin.json", + "sha256": "b89b72bb15d8af650c8d24807501548af5eef233c54f24b2fe4fa68ed82490b6" + }, + { + "path": "commands/test-and-report.md", + "sha256": "540c4788b4cebc1a06c4035bf405a781e69e0a5e2a9635a36fb3cb8a8ac97ab5" + }, + { + "path": "commands/create-messaging-brief.md", + "sha256": "ff3facd9c782196f8e66071ddde82ccf6dc3b3535e5eb77da06b5423bacf6bf4" + }, + { + "path": "commands/generate-channel-copy.md", + "sha256": "99e5aaa4f2f6f08383872e9321c80ed026ec968cc384b461ec08192e4be5a0c3" + }, + { + "path": "skills/voice-guidelines/SKILL.md", + "sha256": "696b10a6a61e1f80537a078950536172c5f28346182d42ae6fa3a36bc2269ebc" + }, + { + "path": "skills/offer-testing/SKILL.md", + "sha256": "f1f98c00cfafe2f93e4712c352b4a45e2a1d60e372dab6ecf947f1908944aa6d" + }, + { + "path": "skills/message-architecture/SKILL.md", + "sha256": "5cb33b4888cee94db0e54cbbb783df8c8aa61a3ed059328737f40cdd3b35687f" + }, + { + "path": "skills/cold-email-personalization/SKILL.md", + "sha256": "14ca3535f55181b4cd51d5e35ecb0622454c3702beb12f2f11f4c66554bfb717" + } + ], + "dirSha256": "513548697dd568a09a97264712fa1bdf4d744bb58768742249549679bdb956bf" + }, + "security": { + "scannedAt": null, + "scannerVersion": null, + "flags": [] + } +} \ No newline at end of file diff --git a/skills/cold-email-personalization/SKILL.md b/skills/cold-email-personalization/SKILL.md new file mode 100644 index 0000000..dfdbfd7 --- /dev/null +++ b/skills/cold-email-personalization/SKILL.md @@ -0,0 +1,43 @@ +--- +name: cold-email-personalization +description: Complete cold email system teaching research-driven personalization, "poke the bear" openers, custom signal hunting, and strict QA. +--- + +# Cold Email Personalization Skill + +## When to Use +- **Cold Outbound Ideation**: Drafting initial emails and subject lines. +- **Personalization**: Mapping research signals to first lines. +- **Sequence Building**: Creating follow-up sequences that rotate value propositions. +- **QA & Optimization**: Reviewing drafts against a strict rubric before sending. + +## Framework +1. **Message Market Fit > Cleverness**: Show you understand the person/company immediately. +2. **Research IS the Personalization**: Custom signals prove you did your homework. +3. **Personalization in the First Line**: Use bracketed variables `{{...}}` or whole-offer strategy. +4. **Tight, Conversational Copy**: Plain text, minimal fluff, 60–120 words. +5. **One Job Per Email**: Single sharp question or CTA. +6. **Earn Replies, Not Just Meetings**: Confirm situation before selling. + +### Core Principles +- **Targeting > Messaging**: Good targeting with bad messaging wastes qualified prospects. +- **Two Paths to Personalization**: Custom Signal Research (Path A) vs. Whole Offer Strategy (Path B). +- **The "Specifically" Line**: "Specifically, it looks like you're trying to sell to {{customer_type}}..." + +## Templates +- **[Research Playbook](assets/research-playbook.md)**: How to find custom signals. +- **[Variable Schema](assets/variable-schema.md)**: Standard variables to capture. +- **[Email Structure](assets/email-structure.md)**: Subject lines, body copy, and "punchiness". +- **[Campaign Types](assets/campaign-types.md)**: Custom Signal, Creative Ideas, Whole Offer, Fallback. +- **[Creative Ideas](assets/creative-ideas.md)**: Generating credible ideas for prospects. +- **[Follow-Up Strategy](assets/follow-up-strategy.md)**: Sequence logic and value rotation. +- **[ICP & Objection Mapping](assets/icp-objection-mapping.md)**: Role-playing the skeptical prospect. +- **[QA Checklist](assets/qa-checklist.md)**: Pre-send verification. +- **[Scoring Rubric](assets/scoring-rubric.md)**: 0-100 quality score. +- **[Examples](assets/examples.md)**: Real campaign examples. + +## Tips +- **Role-Play First**: Always complete the [ICP & Objection Mapping](assets/icp-objection-mapping.md) before writing. +- **Strict QA**: Use the [QA Checklist](assets/qa-checklist.md) to ensure every email meets the 3:1 recipient:sender ratio. +- **Fresh Signals**: Only use signals from the last 90 days. +- **Don't Hallucinate**: If you can't verify a fact, don't use it. diff --git a/skills/message-architecture/SKILL.md b/skills/message-architecture/SKILL.md new file mode 100644 index 0000000..8c2cefd --- /dev/null +++ b/skills/message-architecture/SKILL.md @@ -0,0 +1,30 @@ +--- +name: message-architecture +description: Use when structuring messaging hierarchy, hook banks, and CTA playbooks. +--- + +# Message Architecture Systems Skill + +## When to Use +- Creating new messaging for campaigns, features, or personas. +- Auditing messaging consistency across channels. +- Onboarding new writers or agencies. + +## Framework +1. **Audience & Insight** – persona, pain, desired outcome. +2. **Promise** – primary value proposition + emotional appeal. +3. **Proof** – metrics, customer quotes, product differentiators. +4. **Hooks** – formulas mapped to channels. +5. **CTAs** – stage-specific calls to action + urgency modifiers. + +## Templates +- Message house diagram. +- Hook bank sheet with formulas/examples. +- CTA matrix (persona x funnel stage x CTA). + +## Tips +- Document banned phrases and compliance notes. +- Keep proof points fresh; retire outdated metrics. +- Pair with `voice-guidelines` skill to maintain tone consistency. + +--- diff --git a/skills/offer-testing/SKILL.md b/skills/offer-testing/SKILL.md new file mode 100644 index 0000000..dbbf4c2 --- /dev/null +++ b/skills/offer-testing/SKILL.md @@ -0,0 +1,30 @@ +--- +name: offer-testing +description: Use when designing copy experiments to optimize hooks, offers, and CTAs. +--- + +# Offer Testing Playbooks Skill + +## When to Use +- Planning subject line/CTA/offer tests across email, ads, or landing pages. +- Validating new positioning or pricing language. +- Running copy refresh cycles for campaigns. + +## Framework +1. **Hypothesis** – statement of expected lift + rationale. +2. **Variable Selection** – hook, CTA, body copy, offer framing, proof element. +3. **Segmentation** – define audience splits and holdouts. +4. **Metrics** – primary KPI + guardrails (opens, CTR, CVR, CPL, unsub, spam). +5. **Analysis** – statistical significance (chi-square, z-test) or Bayesian approach. + +## Templates +- Experiment brief (variable, control, variant, KPI, sample size, duration). +- Results report (metric table, significance, insight, next steps). +- Prioritization matrix (ICE/RICE scoring). + +## Tips +- Limit to one variable per test to isolate learnings. +- Ensure minimum sample sizes per channel before declaring winners. +- Log tests and learnings in a shared repository. + +--- diff --git a/skills/voice-guidelines/SKILL.md b/skills/voice-guidelines/SKILL.md new file mode 100644 index 0000000..88ccef0 --- /dev/null +++ b/skills/voice-guidelines/SKILL.md @@ -0,0 +1,30 @@ +--- +name: voice-guidelines +description: Use when enforcing brand tone, style, and localization rules across copy. +--- + +# Voice & Style Guidelines Skill + +## When to Use +- Reviewing copy for tone consistency. +- Localizing copy for new regions or verticals. +- Training new writers or agencies on brand standards. + +## Framework +1. **Voice Pillars** – e.g., confident, empathetic, pragmatic (with examples). +2. **Tone Ladder** – how tone shifts by channel, funnel stage, or audience. +3. **Style Rules** – grammar, punctuation, casing, formatting, banned phrases. +4. **Accessibility** – reading level, inclusive language, jargon policy. +5. **Localization Notes** – regional terminology, regulatory statements, measurement units. + +## Templates +- Do/Don’t example library. +- Localization matrix (region, language, required changes). +- Voice feedback form (issue, recommendation, owner). + +## Tips +- Revisit guidelines quarterly; align with positioning updates. +- Pair with `message-architecture` for consistent structure. +- Track exceptions (e.g., partner co-branding) to avoid conflicting signals. + +---