Digital Marketing Pro
Your agency just signed a 50-brand client. The previous agency left no playbook. Three brands are bleeding budget, two have stale positioning, one is launching in a regulated jurisdiction next month. Where do you start?
Run /digital-marketing-pro:engagement against each brand. Same 12-Part Strategy Flow, same Four Core Documents, same 61-step structure — auditable across the entire portfolio in about an hour per brand on a top-tier model (measured in 2026; your model and plan change the time). No more inconsistent depth between brands. No more “what did the last agency do?” mysteries. No more compliance gaps in regulated jurisdictions.
Open-source AI marketing plugin — 164 skills, 24 specialist agents, EU AI Act Article 50 ready, Cowork team-persistent. Built for marketing agencies, in-house teams running 50–200 brands, and consultancies. Installs on Claude Code (CLI + IDE), Anthropic Cowork, OpenAI Codex, Cursor 2.5+, GitHub Copilot CLI, Google Antigravity 2.0, Hermes Agent, OpenClaw, and Grok + 35+ Agent Skills platforms. Created by Indranil Banerjee · LinkedIn · X.
🆕 Just shipped — v3.42.0 (November 7, 2026): security fixes: Claude Code asks before every step of a live write (post, email, CRM, ads), the send, publish and launch skills start only when you run them, and text from web pages or client files is treated as data and never pasted into a shell command. What’s new · Changelog
# Install — one command (Claude Code 2.1.292 or newer)
claude plugin install digital-marketing-pro --marketplace indranilbanerjee/neels-plugins
Older Claude Code, or inside a session: /plugin marketplace add indranilbanerjee/neels-plugins, then /plugin install digital-marketing-pro@neels-plugins.
If this saves you time, give it a star ⭐ — it’s the single thing that helps other marketers find it.
How it works
Every brand runs the same 12 parts in the same order; Part 5 is the one stop where the client validates. Details: The 12-Part Engagement Methodology.
Try this first
Install, then ask in plain words. Each of these kinds of request reached the right skill in our trigger tests.
| You type |
What happens |
| “set up a new brand” |
The brand profile every skill reads: voice, audience, compliance |
| “build the Q3 campaign plan” |
A multi-channel plan: objectives, channel mix, budget, timeline, KPIs |
| “why did our traffic drop” |
A site-wide SEO audit ranked into a PLAN.md: technical, content, links, local |
| “how do we stack up against X” |
A one-off analysis of 2-5 competitors: positioning, content, SEO, ads, pricing |
| “check this before we publish” |
The scored pre-publish gate: claims, brand voice, compliance; plus an AI-tell scan |
| “how many visitors per variant” |
An A/B test plan computed by script: sample size, days to run, stopping rules |
Safety and approvals
- Writes the plugin sends itself (
/digital-marketing-pro:execute-action) fire only against a single-use approval record for that exact request: the script prepares the record with a preview of the request, the skill approves it after you type yes, and the request must match it byte for byte within 15 minutes. The record proves the approval step ran for this request, once; it cannot prove who typed yes, because the agent runs every command.
- Writes through MCP server tools (Google Ads, Meta, LinkedIn, TikTok, Amazon and other connected servers) are outside that code check. They rely on the skill’s typed-
yes gate and your host’s permission prompt.
- Keep the gate you control. Leave
connector_executor.py --execute and MCP write tools out of your host’s command allowlist so it asks you every time; bypass or auto-approve modes remove that check.
- Autopilot proposes by default. It applies a correction on its own only under a standing approval you approved: scoped, capped per day, at most 30 days. Details in PRIVACY.md.
What it will never do
- Install a package on its own. A missing one prints the exact pinned install command.
- Send, publish or spend without your typed
yes (or a standing rule you approved). Writes it sends itself also need the single-use approval record above.
- Ask for an API key in the chat. Connectors read keys from environment variables.
- Connect a service you did not set up. No MCP server ships enabled. The one hook only makes your host ask before a live write.
- Send your access logs anywhere. The crawler-log report reads them on your machine, prints no full IP address, and downloads only the vendors’ public crawler address lists, when you ask.
- Remove or hide AI watermarks. AI involvement is disclosed, with C2PA provenance where the format supports it.
What it costs
- The plugin is free (MIT-licensed).
- Model usage (tokens). On a subscription (Claude Pro, Max, Team or Enterprise in Claude Code or Cowork; a ChatGPT plan in Codex; a Cursor or Copilot plan; the same goes for other hosts’ plans) the tokens count against your plan’s usage limits instead of being billed per token; check your plan for any optional extra-usage billing. Only pay-as-you-go API-key users pay per token. Context (the skill listing, a skill’s instructions, the documents it reads) uses context window and usage allowance for everyone, and money only for API users. A full engagement is a heavy user;
/usage in Claude Code shows per-model consumption.
- Services you connect with your own keys (ad, email and CRM platforms, SEO and data APIs such as Ahrefs or DataForSEO) are billed by those providers whatever plan you are on, and the money an action moves (an ad budget, a send) is spent on their side. Live writes wait for your typed
yes first.
Get started in 5 minutes (non-developer path)
Are you a marketer, agency owner, or content lead who doesn’t live in a terminal? Here’s the fastest path:
- Open Anthropic Cowork in your browser (no installation, no terminal, no command line). Sign in with your Claude account (Cowork usage counts against your plan).
- Click your profile menu → Settings → Plugins → Add Marketplace. Paste:
indranilbanerjee/neels-plugins
- Find “Digital Marketing Pro” in the list → click Install.
- Type in chat: “Let’s set up a brand for ACME Corp” — Claude will walk you through brand setup (voice, audience, jurisdiction, competitors).
- Then ask: “Run a full marketing engagement for ACME” — and watch ~50–60 strategy documents get produced over the next ~60 minutes.
That’s it. You never touched a command line. Your team Drive will hold the outputs. Re-open Cowork tomorrow and pick up where you left off.
If you’re more technical, see Quick start below for the Claude Code CLI install (one terminal command).
For team usage (agencies running 50+ brands), also run /digital-marketing-pro:cowork-setup once so brand state persists across Cowork sessions via your team’s Google Drive.
Quick start
1. Install on Claude Code (canonical)
One command from your terminal (Claude Code 2.1.292 or newer; it adds the marketplace if needed):
claude plugin install digital-marketing-pro --marketplace indranilbanerjee/neels-plugins
Older Claude Code, or inside a session, takes two steps:
/plugin marketplace add indranilbanerjee/neels-plugins
/plugin install digital-marketing-pro@neels-plugins
/plugin commands work in Claude Code (CLI + IDE at claude.com/code) and Anthropic Cowork. In the standard Claude chat app (browser claude.ai OR the installed Claude Desktop app) plugins still install and run, but management is via the Plugins UI button at the bottom of the chat — not via /plugin slash commands. See the Updating section for the recovery procedure if you accidentally try a slash command in the chat UI.
2. Turn on auto-update (recommended)
Third-party marketplaces have auto-update OFF by default in Claude Code — no banner tells you when a new version ships. Fix it once:
Open /plugin → Marketplaces tab → find neels-plugins → toggle Enable auto-update. Done — future releases pull at session start; /reload-plugins applies mid-session without restart.
3. Set up your first brand
/digital-marketing-pro:brand-setup
Interactive brand profiling — voice, audience, channels, industry, target jurisdictions, competitors, goals. Quick mode (5 questions) or full mode (17 questions). Optional: /digital-marketing-pro:import-guidelines to bulk-load existing brand guidelines, SOPs, or templates.
4. Run a full engagement, or jump straight to a workflow
/digital-marketing-pro:engagement # full 12-Part Strategy Flow (~60 min)
Or jump straight to one workflow:
/digital-marketing-pro:campaign-plan # multi-channel campaign with budget, timeline, KPIs
/digital-marketing-pro:seo-audit # technical + content + E-E-A-T + AI visibility audit
/digital-marketing-pro:content-engine # blog / ad / email / social / landing / video drafts
/digital-marketing-pro:competitor-analysis # multi-dimensional deep-dive
/digital-marketing-pro:performance-report # trends + anomalies + recommendations
/digital-marketing-pro:email-sequence # subject lines, copy, timing, segmentation
/digital-marketing-pro:check # pre-publish quality gate (hallucination + voice + claims)
/digital-marketing-pro:status # unified brand snapshot
/digital-marketing-pro:resume # resume an interrupted long workflow (engagement / campaign-plan / etc.)
/digital-marketing-pro:output-folder # open the user-visible ~/Documents/DigitalMarketingPro/ folder
5. Find your output
~/.claude-marketing/brands/<brand-slug>/
├── profile.json ← brand voice, audience, guardrails, jurisdictions
├── insights.json ← cross-engagement learnings
└── engagements/
└── <engagement-id>/
├── _engagement.json ← state: current part, completed parts, change log
├── living-instruction-file.md ← Living Project Instruction File
├── part-01-client-inputs/ ← Part 1 Stone-vs-Opinion intake
├── part-02-external-research/ ← Part 2 external market research
├── part-03-four-core-documents/ ← Part 3 Four Core Documents (61 steps), v1/ and v2/
├── part-04-competitive-customer-market/ ← Part 4 analyses, v1/ and v2/
├── part-05-client-validation/ ← Part 5 Client Validation Document
├── part-06-v2-reruns/ ← Part 6 selective v2 re-runs
├── part-07-preparation/ ← Part 7 internal operating layer
├── part-08-growth-plan/ ← Part 8 Growth Plan + Yearly Planner
├── part-09-channel-strategy/ ← Part 9 channel-strategy fan-out
├── part-10-execution-artefacts/ ← Part 10 ad copy / post copy / headlines / CTAs
├── part-11-ai-creative-instructions/ ← Part 11 AI creative instructions
└── part-12-continuous-improvement/ ← Part 12 continuous improvement loop
The workspace is ~/.claude-marketing/ unless CLAUDE_MARKETING_HOME or your host’s plugin data folder says otherwise; user-facing deliverables are also copied to ~/Documents/DigitalMarketingPro/ (/digital-marketing-pro:output-folder opens it).
See the Multi-Brand & Agency Guide for the multi-client switching workflow.
Who this is for
| If you’re a… |
Run this |
What you get |
| 🏢 Marketing agency managing 50–200 brands |
/digital-marketing-pro:engagement per brand, then /digital-marketing-pro:cowork-setup for team Drive persistence |
Same 12-Part Strategy Flow audited across every brand. Portfolio health and budget pacing per brand via :agency-dashboard (it adds an AI-cost line only from a /usage output you paste; it cannot read usage itself). |
| 👔 In-house marketing team (B2B SaaS · e-commerce · fintech · healthtech) |
/digital-marketing-pro:engagement once to anchor strategy, then :content-engine + :campaign-plan for ongoing work |
A single canonical strategy doc, monthly stakeholder reports via :performance-report, content + campaigns that tie back to the strategy instead of drifting. |
| 🚀 Marketing automation builder (n8n · Zapier · Make · Pipedream · custom) |
/digital-marketing-pro:doctor to see what’s wired, :execute-action to fire real API calls |
8 verified HTTP connectors executing end-to-end (Slack · HubSpot · Klaviyo · SendGrid · Brevo · Customer.io · Mailchimp · Ahrefs); 28 OAuth connectors via MCP manifest. Stdlib only, no third-party deps. |
| 💼 Solo consultant or freelance marketer |
/digital-marketing-pro:engagement per client |
Same depth on every project. Installs on Codex / Cursor / Copilot CLI / Antigravity if you don’t live in Claude. |
| 📈 Growth team / product marketer |
:funnel-architect → :analytics-insights → :attribution-model → :churn-risk → :cohort-analysis |
Journey design + measurement + retention + churn — all aligned to the strategy document, not isolated outputs. MMM + incrementality testing baked in. |
| 🛡 Compliance-led marketer (EU · UK · India · Brazil · California) |
/digital-marketing-pro:check before publishing anything |
C2PA content provenance, EU AI Act Article 50 disclosure, GDPR + CCPA + DPDPA + LGPD across 16 jurisdictions, deepfake disclosure clauses on every AI creative brief. |
How does this compare?
Other marketing plugins exist, including Anthropic’s own marketing plugin and community SEO skill sets, and each is good at what it covers. What sets Digital Marketing Pro apart:
|
Digital Marketing Pro |
| Scope |
164 skills and 24 specialist agents across strategy, SEO/AEO/GEO, paid media, email, CRM and analytics |
| A methodology |
12-Part Strategy Flow (61 explicit steps) producing the Four Core Documents |
| Agencies |
Per-brand state, brand switching and an agency dashboard |
| Compliance |
EU AI Act Article 50 (C2PA + AI disclosure) and 16 privacy jurisdictions |
| Teams on Cowork |
Persistence through a Drive MCP |
| Real API execution |
8 connectors live, 28 manifest-ready for the MCP path |
| AEO / GEO |
A 6-platform audit, including Google AI Mode |
| Where it runs |
9 native hosts: Claude Code, Cowork, Codex, Cursor, Copilot CLI, Antigravity, Hermes, OpenClaw, Grok |
| Live-write safety |
Single-use approval record for writes it sends itself; MCP writes: typed yes + host prompt |
| Tests |
1099 stdlib unittest tests |
| License |
MIT: no telemetry, no seats |
Why Digital Marketing Pro
Most AI marketing tools generate isolated outputs — a campaign brief here, an email there. No canonical sequence, no shared state, no enforced structure. Result: inconsistent depth, missed dependencies, outputs that don’t compound.
DM Pro runs every brand through the same 12 parts, producing the same files in the same order, with explicit dependency rules between them. That’s the whole product. Everything else — the 164 skills, 24 agents, 2026 compliance updates, Cowork persistence — exists to make that 12-Part Flow ship cleanly across real marketing operations.
| What this gives you that ad-hoc prompts don’t |
Why it matters |
| Canonical 12-Part Strategy Flow producing the Four Core Documents (61 explicit steps) |
Every engagement looks the same, so handoffs work and quality is auditable |
| Two-Views Model (v1 unbiased + v2 client-validated) |
You never lose the original market view when the client pushes back |
| Decision Matrix — maps validation responses to re-runs |
Stops over-running (wasted hours) and under-running (broken strategy) |
| Living Project Instruction File — single source of truth per engagement |
All skills read it first; corrections propagate automatically |
| EU AI Act Article 50 compliance built in |
C2PA provenance signing, deepfake disclosure, final Article 50 Guidelines + Code of Practice (10 June 2026) in compliance |
| 6-platform AEO/GEO audit (incl. Google AI Mode) |
Treats AI Mode as a distinct surface from AI Overviews |
What you get in 60 minutes
Run /digital-marketing-pro:engagement and the plugin produces a full brand-strategy engagement in roughly 60 minutes on a top-tier model (measured on Opus 4.8) — ~50–60 canonical files organized by part:
- Part 1 — Stone-vs-Opinion intake (what the client knows for certain vs what they believe)
- Part 2 — External market research (unbiased, no client docs)
- Part 3 — Four Core Documents — 61 explicit steps across Business & SBU Analysis, Segmentation Framework, Brand Positioning & Communications, DMFlow
- Part 4 — Competitive + Customer + Market analysis (4 unbiased docs)
- Part 5 — Client Validation Document — the one true stop
- Part 6 — Selective v2 re-runs per Decision Matrix
- Part 7 — Preparation documents (campaign architecture, KPI tree, content pillars, approval chains)
- Part 8 — Growth Plan + 12-month Yearly Planner (the flagship deliverable)
- Part 9 — Channel-strategy fan-out (up to 17 channel docs in 7 families)
- Part 10 — Execution artefacts (ad copy, post copy, headlines, CTAs)
- Part 11 — AI creative briefs (with current image- and video-model guidance and C2PA + deepfake-disclosure clauses)
- Part 12 — Continuous improvement loop
Cost: the plugin is MIT-licensed and free. A full 12-part engagement writes about 50–60 documents, so it uses a lot of model tokens: on a subscription that counts against your plan’s usage limits, and only pay-as-you-go API-key users are billed per token (What it costs).
Real workflows you’d actually run
🆕 New-client onboarding (agency, week 1)
/digital-marketing-pro:brand-setup "ACME Corp" # interactive: voice, audience, channels, jurisdiction
/digital-marketing-pro:competitor-analysis # multi-dimensional deep-dive on top 5 competitors
/digital-marketing-pro:engagement # full 12-Part Strategy Flow (~60 min on Opus-class)
/digital-marketing-pro:check engagements/<id>/part-03-four-core-documents/v1/*.md # pre-publish gate before client review
Output: ~50–60 canonical files. Time saved: ~3 weeks of senior-strategist labor.
📊 Quarterly business review (in-house, last week of quarter)
/digital-marketing-pro:performance-report --period=Q2-2026
/digital-marketing-pro:attribution-report --period=Q2-2026 --model=data-driven
/digital-marketing-pro:competitor-monitor --since=2026-04-01
/digital-marketing-pro:continuous-improvement-loop --quarter=Q2-2026
Output: stakeholder-ready Q2 review with anomalies, attribution shift, competitor moves, and next-quarter recommendations.
🎯 SEO sprint (any audience, 1 week)
/digital-marketing-pro:seo-plan # 4-pillar scorecard; weakest pillar drives the theme
/digital-marketing-pro:keyword-cluster seeds.csv # SERP-overlap clustering into pillar+spokes
/digital-marketing-pro:backlink-gap acme.com competitor1.com competitor2.com
/digital-marketing-pro:content-engine # draft the top 3 pillar pages
/digital-marketing-pro:check drafts/*.md # hallucination + brand voice + claims gate
/digital-marketing-pro:seo-drift baseline.csv current.csv # 30 days later, what moved
🤖 Marketing automation flow (builders)
/digital-marketing-pro:doctor # which actions are live vs need connector setup
/digital-marketing-pro:execute-action --action diagnostic --brand acme --execute # GA4 + GSC tag/consent check (MCP path: returns the manifest)
/digital-marketing-pro:execute-action --action audit-current --brand acme --execute # current SEO state (Ahrefs over HTTP; GSC via MCP)
/digital-marketing-pro:execute-action --action enable-automation --brand acme --execute # email automation flow: prepares a preview + approval record
# after you type yes: approval-manager.py --brand acme --action approve --id <id>, then re-run with --approval-id <id>
Output: HTTP calls for the 8 executable connectors (Slack, HubSpot, Klaviyo, SendGrid, Brevo, Customer.io, Mailchimp, Ahrefs), MCP manifests for the rest; each write fires against a single-use approval record for that exact request and is logged at ~/.claude-marketing/brands/{brand}/executions/. See Safety and approvals.
🛡 Pre-publish compliance gate (every campaign)
/digital-marketing-pro:check campaign.md --full # hallucination + voice + claims + jurisdictions
/digital-marketing-pro:c2pa-metadata hero.png # sign image with provenance for EU Article 50
🎨 AI creative brief with EU disclosure (every AI-generated asset)
/digital-marketing-pro:ad-creative # ad concepts + copy with EU/FTC disclosure clauses
/digital-marketing-pro:influencer-creator # FTC + EU deepfake clauses baked in
Supported surfaces (v3.42.0)
| Platform |
Install command |
Manifest path |
Status |
| Claude Code CLI + IDE extensions |
claude plugin install digital-marketing-pro --marketplace indranilbanerjee/neels-plugins (2.1.292+), or /plugin marketplace add indranilbanerjee/neels-plugins then /plugin install digital-marketing-pro@neels-plugins |
.claude-plugin/plugin.json |
Full support (canonical) |
| Anthropic Cowork |
Plugins UI → Add marketplace → indranilbanerjee/neels-plugins → Install |
same .claude-plugin/ files |
Full support — no /plugin slash commands in Cowork (UI-only) |
| OpenAI Codex CLI + IDE + App |
codex plugin marketplace add indranilbanerjee/neels-plugins then codex plugin add digital-marketing-pro@neels-plugins |
.codex-plugin/plugin.json (published OpenAI schema) |
Full skills + MCP support |
| Cursor 2.5+ |
In any Cursor Agent chat: /add-plugin digital-marketing-pro@https://github.com/indranilbanerjee/digital-marketing-pro |
.cursor-plugin/plugin.json (published Cursor JSON Schema) |
Full skills + agents + commands support |
| GitHub Copilot CLI |
copilot plugin marketplace add indranilbanerjee/neels-plugins then copilot plugin install digital-marketing-pro@neels-plugins |
.github/plugin/plugin.json (Copilot CLI also recognizes .claude-plugin/plugin.json as fallback) |
Full skills + MCP support; subagents need .agent.md extension (open issue); skills run as /skill-name or by natural language |
| Google Antigravity 2.0 CLI + IDE |
agy plugin install https://github.com/indranilbanerjee/digital-marketing-pro |
gemini-extension.json (at repo root, per Google’s reference pattern) |
Full skills + hooks support; subagents need /agent CLI spawning; slash commands fold into skills via agy plugin import gemini |
| Hermes Agent (Nous Research) — Desktop + CLI on macOS / Windows / Linux |
hermes plugins install indranilbanerjee/digital-marketing-pro |
plugin.yaml + __init__.py at repo root (Hermes native spec) |
Native plugin — adapter walks skills/ at register time and exposes all 164 skills via ctx.register_skill(). Targets Hermes Desktop v0.15.2+ (public preview June 2 2026). |
| OpenClaw (formerly Clawdbot / Moltbot) |
openclaw plugins install git:github.com/indranilbanerjee/digital-marketing-pro |
openclaw.plugin.json at repo root (also auto-detects .claude-plugin/plugin.json as Claude-compatible bundle) |
Native plugin via openclaw.plugin.json; skills field points at ./skills. Also installable via ClawHub marketplace (submission pending). |
| Grok (xAI Build CLI) |
grok plugin install indranilbanerjee/digital-marketing-pro — or grok plugin marketplace add indranilbanerjee/neels-plugins then grok plugin install digital-marketing-pro (append --trust to skip the install confirmation) |
.grok-plugin/plugin.json + .grok-plugin/marketplace.json (Grok Build also reads the Claude Code manifests for compatibility; the native pair is the first-class lane) |
Full skills support |
Why this works: Agent Skills became an open standard in December 2025 (donated to the Agentic AI Foundation; adopted by 41+ agent products by June 2026 — see “Works on 40+ agent harnesses” below). All 164 SKILL.md files in DM Pro are platform-portable as written. The sibling manifests are thin platform-specific wrappers around the same skills/ directory — no skill duplication, no maintenance fork. The pattern is borrowed from Google’s reference repo gemini-cli-extensions/data-agent-kit-starter-pack.
Recommended Claude Code version: current. Claude Code has no plugin-level minimum-version field — requiredMinimumVersion is a managed setting that pins the host version for everyone, so a plugin cannot enforce it (DMP declared it in plugin.json from v3.12.0 to v3.31.1, where Claude Code silently stripped it; corrected in v3.32.0). Teams that want a floor set it in their own settings.json — see settings.json.example.
Works on 40+ agent harnesses (via the Agent Skills open standard)
Beyond the 9 surfaces above where we ship a native manifest, DMP’s 164 SKILL.md files work out-of-the-box on any agent that adopted the Agent Skills open standard (Anthropic-published Dec 2025, 41+ adopters as of June 2026). On each platform below, copy our skill folders (the folders inside skills/) into the folder it reads and all 164 marketing skills are discoverable. No platform-specific manifest needed.
Platforms that have adopted the Agent Skills standard. Where a row gives an install hint, it was checked against that tool’s own documentation in October 2026; most of these tools read skills from a folder, so the hint names the folder to copy the skill folders into:
| Platform |
Vendor |
Install hint |
| Goose |
Block (Square) |
Copy the skill folders into ~/.agents/skills/ (all projects) or .agents/skills/ (one project) · docs |
| OpenHands |
Open Hands (cloud agents) |
Copy the skill folders into the project’s .agents/skills/ · docs |
| OpenCode |
sst |
Copy the skill folders into .opencode/skills/ (project) or ~/.config/opencode/skills/ (global); .agents/skills/ and .claude/skills/ are read too · docs |
| Junie |
JetBrains |
Copy the skill folders into .junie/skills/ (project) or ~/.junie/skills/; Junie CLI also reads .agents/skills/ in a trusted project, and /skills re-scans · docs |
| Gemini CLI |
Google |
Copy the skill folders into ~/.gemini/skills/ (or ~/.agents/skills/) for all projects, or .gemini/skills/ for one · docs |
| Roo Code |
Roo Code Inc. |
Copy the skill folders into .roo/skills/ (project) or ~/.roo/skills/ (global); .agents/skills/ works too · docs |
| Cline / Windsurf |
open-source VS Code agents |
Cline: copy into .cline/skills/ or ~/.cline/skills/ (docs). Windsurf: copy into .devin/skills/ (the legacy .windsurf/skills/ is still read) (docs) |
| Kiro |
Kiro |
Spec-driven dev with Agent Skills support |
| Amp |
Sourcegraph |
Copy the skill folders into .agents/skills/ (project) or ~/.config/agents/skills/ (this machine) · docs |
| Letta |
Letta |
Copy the skill folders into .agents/skills/ (project) or ~/.letta/skills/ (all agents on this machine) · docs |
| Mux |
Coder |
Browser-based parallel cloud agents |
| Factory |
Factory |
“Droid” agents read Agent Skills bundles |
| Workshop |
Workshop |
Multi-LLM cross-platform agent |
| Tabnine |
Tabnine |
Enterprise context-aware AI agent |
| Emdash |
General Action |
Parallel git-worktree agents |
| Superconductor |
Superconductor |
Multiplayer cloud agents |
| Ona |
Ona |
Background cloud-agent fleet |
| Mistral Vibe |
Mistral AI |
Copy the skill folders into .vibe/skills/ (project) or ~/.vibe/skills/; .agents/skills/ is read too · docs |
| VT Code |
open-source |
LLM-native code agent |
| Qodo |
Qodo |
Code integrity agent |
| Piebald |
Piebald |
Desktop agentic dev |
| Autohand Code CLI |
Autohand |
ReAct terminal agent |
| pi |
open-source |
Minimal terminal harness |
| Command Code |
Command Code |
Coding-taste-learning agent |
| TRAE |
ByteDance |
Adaptive AI IDE |
| Firebender |
Firebender |
Android-native agent |
| bub |
Bub |
Channel-native agent framework |
| fast-agent |
evalstate |
ACPX + Skills development |
| nanobot |
HKUDS |
Ultra-light personal agent (Slack / Discord / Telegram / WeChat) |
| Vita |
Vita |
Virtual-desktop autonomous workers |
| Snowflake Cortex Code |
Snowflake |
Data-platform agent |
| Databricks Genie Code |
Databricks |
Data-engineering agent |
| Laravel Boost |
Laravel |
Laravel-specific agent skills layer |
| Spring AI |
Spring |
Java/Spring AI applications |
| Agentman |
Agentman |
Healthcare revenue-cycle agents |
| Google AI Edge Gallery |
Google |
On-device mobile LLM agent |
Quick test on any of these platforms:
# 1. Clone the repository
git clone --depth=1 https://github.com/indranilbanerjee/digital-marketing-pro.git
# 2. Copy the folders inside ./digital-marketing-pro/skills/ into your tool's skills folder (see the table above)
# 3. Try: "Run a competitor analysis on stripe.com"
# Your agent should pick the competitor-analysis skill.
Why we don’t ship per-platform manifests for these: the Agent Skills standard says agents discover by walking a directory tree for SKILL.md files — no manifest required. Shipping 35 extra wrapper manifests would create maintenance overhead with zero added value.
If you run into a platform-specific install snag, file a GitHub issue — we’ll add platform-specific docs as users report patterns.
The 12-Part Engagement Methodology
| Part |
Name |
Output |
| 1 |
Client Inputs |
Stone vs Opinion intake (what client knows for certain vs what they believe) |
| 2 |
External Research |
Unbiased market research (no client docs used) |
| 3 |
Four Core Documents |
61 explicit steps — Business & SBU (18), Segmentation (15), Brand Positioning (19), DMFlow (9) |
| 4 |
Competitive + Customer + Market |
4 unbiased analysis documents (4.1–4.4) |
| 5 |
Client Validation Document |
The one true stop — client accepts/rejects/edits each finding |
| 6 |
Selective v2 Re-runs |
Subset of Part 3 + Part 4 docs re-run per the Decision Matrix |
| 7 |
Preparation Documents |
Internal operating layer (campaign architecture, KPI tree, content pillars, asset inventory, approval chains) |
| 8 |
Growth Plan + Yearly Planner |
The flagship 11-section client-facing strategy + 12-month operational calendar |
| 9 |
Channel Strategy Fan-out |
Up to 17 channel docs grouped into 7 families |
| 10 |
Execution Artefacts |
Ad copy, post copy, headlines, CTAs |
| 11 |
AI Creative Instructions |
Visual asset briefs with C2PA + EU Article 50 clauses |
| 12 |
Continuous Improvement Loop |
Quarterly briefs feeding signals back into product/offering decisions |
Key architectural concepts:
- Two-Views Model — Every engagement carries v1 (unbiased market view) and v2 (client-validated view) after Part 5. Operating decisions reference v2; ideation references both. v1 is never deleted.
- Stone vs Opinion — Every fact captured at intake is tagged with confidence. Stone = client knows for certain. Opinion = client believes (becomes a research question, not ground truth).
- Decision Matrix — Maps client validation responses to which v1 documents need v2 re-runs. Prevents over- and under-re-running.
- Update-Back Rule — Live operations surface corrections → source documents get versioned (v2.1, v2.2 …) → Living Project Instruction File propagates the change to all downstream skills.
- Living Project Instruction File — Single source of truth per engagement. All skills read it first.
15+ strategic-framework reference documents in skills/context-engine/ support the methodology (Five Digital Markets, Channel Families, In-Market vs Out-Market, Multi-Dimensional Decision Framework, Unit Economics, Actionable Persona Format, B2B Decision-Making Unit, Three-Scenario Forecasting, 30/60/90-Day Framework, Reporting Cadence, Fixed vs Variable Budget, Competitor 3-Question Output, India Market Context, and more).
What’s new
v3.42.0 — the host asks before every live write, outside text stays data (November 7, 2026)
A security review of the plugin found that a live write (a Slack post, an email send, a CRM or ads change) could run without your host ever asking: 14 skills and commands pre-approved every shell command, and the approve step is a command the model runs itself. Now no skill pre-approves the shell beyond named local scripts, and on Claude Code the plugin’s one hook makes the host ask before each prepare, approve and fire step, even when you saved “don’t ask again”. The send, publish, launch and sync skills start only when you run them as commands. Every agent, skill and command now says that fetched, uploaded or pasted text is data, never instructions, and JSON that may hold outside text goes through a file (--data "@{data_file}"), not a single-quoted shell argument. See CHANGELOG.md and UPGRADE-GUIDE.md.
v3.41.0 — a zero-context review, what AI crawlers actually did, and a watch list for ad products (October 29, 2026)
Whoever writes a plan reads past what it assumes; the client doesn’t. /digital-marketing-pro:check --goldfish packs the deliverable, its brief and the brand profile into one file and nothing else, flags any money figure the brief never states and any benchmark rate with no source, and gives a reviewer a prompt that names only that file. Client proposals, reports, plans and summaries now end with that step. The agent-readiness audit can also read the site’s own access logs: crawler-log-report.py counts what the OpenAI, Anthropic, Perplexity, Google, Microsoft and Apple crawlers fetched, and checks each request against the addresses the vendor publishes, so a fake GPTBot shows up as spoofed. The logs never leave your machine, and the report prints no full IP address. New ad products (Meta’s Andromeda, Google’s campaign total budgets and AI Brief, TikTok’s October announcements, X’s ad-text rules and more) sit on a dated watch list that says whether each was read at its source or only reported, so a plan doesn’t budget for something that isn’t live. TikTok’s developer reference settles the caption: 2,200 characters for a video; the 4,000 quoted elsewhere is a photo post’s description. 1063 tests.
v3.40.0 — an audit release: safer keys and brands, a gate that blocks, facts re-read at source (October 28, 2026)
Outside audits and customer-journey tests on an installed copy of 3.39.1 found real problems, and this release fixes them. A Mailchimp key with a crafted ending could send itself to another host; now every connector key goes only to its own vendor, and no key, whether a connector’s or a model provider’s, follows a redirect. Creating a brand that already existed replaced its profile with a blank one, and brand names in Hindi, Chinese or emoji all shared one folder; both are now refused or kept apart, and no file a skill writes can land outside its brand. The quality gate never applied a brand’s own settings; it does now, and it blocks a draft with a critical issue. Audits end with how you’ll know each recommendation worked (a metric, a direction and a date), and the continuous-improvement loop checks them when they fall due. connector-status.py --action probe proves a key works with one read-only request. AI-visibility numbers come with their formula, and social-platform limits now live in one sourced file. Platform facts were re-read at their source: Instagram’s 5-hashtag cap, Google Ads’ data-driven default, Privacy Sandbox’s retirement, YouTube’s Shorts and monetisation rules, New York’s synthetic-performer law and more. Some behaviours are stricter, so read the upgrade guide before you upgrade a wrapper. 1027 tests.
v3.39.1 — offline checks read the deprecations record too (October 27, 2026)
When the model router can’t reach a provider’s live list, it checks a model you name against what ships with the plugin. Until now that meant only the model registry, not the deprecations page record that sits right beside it. So, offline, a model the vendor’s page retires could slip through: OpenAI’s o3, which isn’t in the registry, or a Google model switched off in June. A model whose requests Google now hands to a newer one got no word about it either. The router now checks both and acts on the stricter, saying which one it followed. A plugin calling Google through Vertex AI, which retires models on its own timetable, can say the record doesn’t apply. A provider list cut off at the page limit is now refused rather than used short, since a short list would make a model look withdrawn. 882 tests.
Older releases are in CHANGELOG.md.
How the SEO skills chain together
Most SEO work uses 3-5 skills in sequence rather than one mega-skill. The plugin is designed so that each skill produces numbered intermediate files (01-...md, 02-...md, …, PLAN.md) under ${CLAUDE_PLUGIN_DATA}/{brand}/seo/{workflow}/{date}/ — downstream skills read those numbered files, not the endpoint, so you can re-run any single step without redoing the whole chain.
Agency onboarding workflow (week 1 of a new client engagement):
1. /digital-marketing-pro:brand-setup
2. /digital-marketing-pro:competitor-analysis ← picks the right competitors for everything downstream
3. Run all in parallel:
/digital-marketing-pro:tech-seo-audit ← baseline technical health
/digital-marketing-pro:aeo-audit ← baseline AI visibility
/digital-marketing-pro:backlink-gap ← link prospects (needs competitors from step 2)
/digital-marketing-pro:gsc-ai-performance ← GSC AI Performance Report (3 Jun 2026)
4. /digital-marketing-pro:keyword-cluster ← pillar+spokes architecture from aeo-audit content gaps
5. /digital-marketing-pro:seo-plan ← DISPATCHER — reads all of the above, scores 4 pillars,
the weakest pillar drives the lead theme of Q1's roadmap
Quarterly review workflow:
1. /digital-marketing-pro:gsc-ai-performance ← fresh GSC AI export
2. /digital-marketing-pro:seo-drift ← compare this quarter vs last (auto-classifies gainers,
losers, reshuffles, new keys, lost keys)
3. Branch by finding:
- High decline → /digital-marketing-pro:seo-audit + /digital-marketing-pro:content-decay-scan
- High reshuffle → /digital-marketing-pro:aeo-geo (intent realignment)
- High growth → /digital-marketing-pro:content-engine (amplification briefs)
4. /digital-marketing-pro:seo-plan ← re-run dispatcher with fresh inputs;
lead theme may shift to a different pillar
Content production workflow:
1. /digital-marketing-pro:keyword-cluster ← from your seed list
2. /digital-marketing-pro:content-brief ← per pillar from the cluster plan
3. /digital-marketing-pro:content-engine ← drafts with brand voice + fact-check + humanize + SEO checklist
4. /digital-marketing-pro:check ← pre-publish gate (hallucination + brand voice + structure)
5. /digital-marketing-pro:publish-blog ← push to CMS
6. /digital-marketing-pro:c2pa-metadata ← if EU markets are targeted and AI images accompany
Backlink campaign workflow:
1. /digital-marketing-pro:competitor-analysis
2. /digital-marketing-pro:backlink-gap ← gap-vs-competitors with link-prospect priority scoring
3. /digital-marketing-pro:digital-pr ← consumes the prospect shortlist + outreach templates
4. /digital-marketing-pro:pr-pitch ← drafts individual pitches per prospect
Each skill has a quality scorecard that must pass before its PLAN.md is declared ready, and every heavy skill carries a Tips & caveats section with the common pitfalls. The seo-plan dispatcher uses Confirm-Then-Dispatch — it never silently re-runs expensive specialists, always asking explicitly with cost estimate before fanning out.
Architecture — what’s actually in the box
24 specialist agents
Marketing Strategist · Brand Guardian · Content Creator · Email Specialist · Social Media Manager · PR Outreach · SEO Specialist · CRO Specialist · Analytics Analyst · Marketing Scientist · Market Intelligence · Influencer Manager · CRM Manager · Growth Engineer · Journey Orchestrator · Agency Operations · Performance Monitor · Quality Assurance · Memory Manager · Execution Coordinator · Intelligence Curator · Localization Specialist · Media Buyer · Competitive Intel
Each agent has scoped responsibilities, explicit input/output contracts, and reads the Living Project Instruction File before acting.
164 skills
Skills are invoked by description match through the Skill tool, addressable as /digital-marketing-pro:<skill-name> from chat. Coverage: brand setup, content production (blog / ad / email / social / landing / video / PR / case study), SEO / AEO / GEO audits (6 platforms incl. Google AI Mode), competitor monitoring, campaign planning, channel-specific strategies, attribution, churn risk, lifecycle journeys, intelligence reports, eval framework, knowledge management, multi-brand operations, regional configuration, C2PA content provenance, Cowork+Drive team persistence.
Slash commands
Every skill answers to /digital-marketing-pro:<skill-name>. Five more are standalone commands with no same-named skill: engagement, resume, output-folder, doctor and execute-action. v3.33.0 folded every command that duplicated a same-named skill into that skill, so those slash names work exactly as before and each now loads once instead of twice. The most-used:
| Slash name |
What it does |
/digital-marketing-pro:brand-setup |
Set up a new brand profile (voice, audience, competitors, compliance) |
/digital-marketing-pro:engagement |
Run the full 12-Part Strategy Flow |
/digital-marketing-pro:campaign-plan |
Generate a multi-channel campaign plan with budget, timeline, KPIs |
/digital-marketing-pro:seo-audit |
Comprehensive SEO audit — technical, on-page, content, E-E-A-T, AI visibility |
/digital-marketing-pro:content-engine |
Draft blog, ad copy, emails, social, landing pages, video scripts |
/digital-marketing-pro:performance-report |
Performance report with trends, anomaly detection, recommendations |
/digital-marketing-pro:competitor-analysis |
Multi-dimensional competitive analysis (content, SEO, ads, social, pricing) |
/digital-marketing-pro:email-sequence |
Complete email sequences (subject lines, copy, timing, segmentation) |
/digital-marketing-pro:check |
Pre-publish quality gate (hallucination + brand voice + structure + claims) |
/digital-marketing-pro:status |
Unified brand snapshot (profile, engagements, insights, compliance) |
/digital-marketing-pro:resume |
Resume an interrupted long workflow from the last checkpoint |
/digital-marketing-pro:output-folder |
Print + open the visible output folder for a brand |
/digital-marketing-pro:doctor |
Per-action readiness diagnostic (which campaign-audit / launch-campaign actions are live vs need connector setup) |
/digital-marketing-pro:execute-action |
Actually fire an action against its real API (stdlib urllib, no third-party deps). 8 verified connectors execute end-to-end; 28 OAuth-only or MCP-only connectors fall back to the MCP path with the manifest still returned. |
/digital-marketing-pro:cowork-setup |
(v3.12.0) One-shot Cowork team setup — wires DMP through a Drive MCP so brand state survives across Cowork sessions |
/digital-marketing-pro:keyword-cluster |
Pillar + spokes content cluster from seed keywords with SERP-overlap clustering and 4-gate quality scorecard |
/digital-marketing-pro:backlink-gap |
Competitor backlink gap audit with priority scoring (DR + overlap + traffic + topical) |
/digital-marketing-pro:seo-drift |
Snapshot-vs-snapshot drift with auto-classification (growth/decline/reshuffle/stable/new/lost) |
Every other skill works the same way — :competitor-monitor, :churn-risk, :autopilot-status, :agency-dashboard, :aeo-audit, :geo-monitor, :c2pa-metadata, :client-onboarding, :journey-design … see /digital-marketing-pro:help after install for the full list, or browse skills/ in the repo.
101 Python scripts (optional)
Plugin works fully without Python — all marketing knowledge, frameworks, agent capabilities, and skills work out of the box via the 178 reference knowledge files.
| Mode |
Size |
Adds |
| Knowledge-only (default) |
0 MB |
All 164 skills + 24 agents + 178 reference files |
Lite (python scripts/setup.py --install lite) |
~15 MB |
Brand-voice scoring, content quality scoring, readability analysis |
Full (python -m pip install -r scripts/requirements.txt, exact pins) |
~50 MB |
Competitor scraping, QR generation, AI visibility API checking, GEO tracking, C2PA signing |
35 HTTP MCP connectors
Ready-to-copy entries in .mcp.json.connectors-reference: first-party servers such as Notion · Slack · Canva · Figma · HubSpot · Amplitude · Ahrefs · SimilarWeb · Klaviyo · Google Calendar · Gmail · Stripe · Asana · Webflow, plus official ad-platform, image and video, and aggregator MCPs. .mcp.json.example adds 3 vendor npx entries.
These are an opt-in catalog — no .mcp.json ships (it is gitignored), so nothing auto-connects; enable only the ones you need. All HTTP, all Cowork-compatible. For services without first-party HTTP MCPs (Google Sheets, Drive, Salesforce, etc.), see .mcp.json.connectors-reference for Pipedream / Composio / Zapier / Make.com aggregator paths.
Connector setup offers only vendor-verified, version-pinned packages: .mcp.json.example has runnable npx entries only for packages the vendor itself publishes on npm (Pinecone, DataForSEO, DeepL), pinned to an exact version and checked weekly against the npm registry. Every other service uses the vendor’s hosted HTTP server or is set up from the vendor’s own docs; DMP prints no npx command for it. See CONNECTORS.md and Integrations Guide.
Resumable workflows and a visible output folder
Two user-team complaints from the v3.7.5 cycle drove this release: “dm pro is taking too long to process” (the 60-minute engagement that breaks midway loses 30+ minutes of work on restart) and the general “where did my 50 deliverable files save?” confusion (everything was landing under the Windows-hidden ~/.claude-marketing/ dotfolder).
Fix 1 — Resumable workflows. Every long-running DMP workflow now writes per-part checkpoints to disk so an interrupted session can resume from the next un-checkpointed part instead of restarting from Part 1. Covered workflows: engagement (12-Part Strategy Flow), campaign-plan, content-engine, seo-audit, competitor-analysis, campaign-audit (v3.7.5), launch-campaign (v3.7.5), plus a custom slot for any other long flow. Resume with:
/digital-marketing-pro:resume # auto-pick latest in-progress run
/digital-marketing-pro:resume engagement # filter to a workflow
/digital-marketing-pro:resume engagement <run-id> # pick a specific run
Fix 2 — Visible output folder. Every artifact a workflow produces is now copied to TWO locations: the internal tracking copy under ~/.claude-marketing/brands/{brand}/output/{workflow}/... (system-of-record), and a user-visible published copy under ~/Documents/DigitalMarketingPro/{brand}/{workflow}/{YYYY-MM}/{filename} (visible in Windows Explorer / macOS Finder by default; on Windows the script uses the Documents folder Explorer shows, including one OneDrive has moved). Override the visible root with DIGITAL_MARKETING_PRO_PUBLISH_DIR=/path (e.g. a Dropbox share for the team). Reveal the folder any time with:
/digital-marketing-pro:output-folder # opens ~/Documents/DigitalMarketingPro/{brand}/
/digital-marketing-pro:output-folder <brand> <workflow> # drill down
Implementation: scripts/checkpoint-manager.py (per-step storage + atomic writes, stdlib only) + scripts/output-publisher.py (dual-copy publish + where + open subcommands: internal tracking copy + user-visible published copy). The engagement/checkpoint state machine ships tests in tests/test_engagement_state.py + tests/test_checkpoint_roundtrip.py; a fuller 5-scenario end-to-end simulation (clean 12-part run / interrupt-resume / parallel workflows / quality-gate fail / all-workflows-accepted) was used during development but is a dev tool, not shipped in the repo.
Connector-aware action resolver
The campaign-audit and launch-campaign skills depend on 14 actions that map to real marketing APIs (Google Ads, Meta Marketing, LinkedIn, TikTok, HubSpot, Salesforce, Klaviyo, Mailchimp, Customer.io, Gmail, Cision, Muckrack, Slack, Google Calendar, Ahrefs, Similarweb, SEMrush, Google Search Console). v3.7.5–v3.7.7 shipped these actions as honest stubs that always returned status: stub_implementation regardless of which connectors the user had. v3.7.10 introduces a resolver that probes the live state and resolves each action to one of three modes per call:
| mode |
what it means |
real |
runs end-to-end with no external API (currently only arm-watchdog which writes a watchdog config to ~/.claude-marketing/brands/{brand}/watchdogs/) |
manifest_ready |
a matching connector is configured — the response includes the exact HTTP request manifest (method, URL, headers, body template, auth pattern) for the orchestrator (Claude via MCP) to execute. Write/launch ops set approval_required: true. |
stub_unconfigured |
no matching connector is configured — the response includes the manual fallback PLUS copy-paste .mcp.json snippet, env-var list, and a Cowork-compatibility note |
Check what’s live in your environment any time:
/digital-marketing-pro:doctor # full readiness table
/digital-marketing-pro:doctor --summary # one-line counts
/digital-marketing-pro:doctor --action inventory --channel google_ads # drill in
Test coverage: the resolver’s action layer ships tests in tests/test_connector_resolver.py. A fuller 27-scenario development harness (14 actions × unconfigured / configured / local-execution variants) was used while building this out, but it is a dev tool and is not shipped in the repo.
Implementation: scripts/_connector_registry.py (catalog of 36 connectors, 11 categories, is_connector_configured() probe) + scripts/connector_resolver.py (ACTION_SPECS map + per-action manifest builders + local executors) + scripts/action-doctor.py (the doctor command’s underlying script).
v3.7.11 — actions can actually fire HTTP requests from Python
The v3.7.10 resolver returned a manifest of “what would be sent.” v3.7.11 adds scripts/connector_executor.py (stdlib urllib.request, no third-party deps) that takes that manifest and actually executes the request against the real API. Public CLI: /digital-marketing-pro:execute-action.
Executes end-to-end from Python (8 connectors, verified vendor docs):
| Connector |
Env var |
What it can fire |
| Slack |
SLACK_BOT_TOKEN |
POST chat.postMessage (with body.ok post-check) |
| HubSpot |
HUBSPOT_PRIVATE_APP_TOKEN |
GET /automation/v4/flows, POST /marketing/v3/campaigns |
| Klaviyo |
KLAVIYO_PRIVATE_KEY |
GET /api/flows, PATCH /api/flows/{id} (vnd.api+json) |
| SendGrid |
SENDGRID_API_KEY |
POST /v3/mail/send (202 success) |
| Brevo |
BREVO_API_KEY |
GET /v3/emailCampaigns (read only, lowercase api-key: header; sends go through the Brevo MCP) |
| Customer.io |
CUSTOMERIO_APP_API_KEY |
GET /v1/campaigns (read only; sends go through the Customer.io MCP) |
| Mailchimp |
MAILCHIMP_API_KEY |
GET /3.0/automations (Basic auth, dc from suffix) |
| Ahrefs |
AHREFS_API_KEY |
GET /v3/site-explorer/metrics |
Requires the MCP path (28 OAuth-only or MCP-only connectors): Google Ads, Meta Marketing, LinkedIn Marketing, LinkedIn Publishing, TikTok Ads, Twitter/X, Gmail, Google Calendar, Google Analytics, Google Search Console, Meta Graph, Salesforce, Pipedrive, Zoho CRM, Buffer, Hootsuite, Cision, Muckrack, Amplitude, Similarweb, SEMrush, Moz, Intercom, Canva, Figma, plus the official ad-platform MCP servers Meta Ads AI Connectors, Google Ads MCP (read-only) and Amazon Ads MCP. For all of these, the resolver still returns manifest_ready so you can see the exact HTTP shape Claude’s MCP tool will send — Python just can’t execute the OAuth flow itself.
Safety gates (updated in v3.35.0): read ops auto-execute with --execute; write ops need a matching approved, single-use approval record (--approval-id), see Safety and approvals; missing env vars block with setup_hint_credential; unresolved {VAR} placeholders block before the request fires. Every fired call logs to ~/.claude-marketing/brands/{brand}/executions/.
Test coverage: end-to-end HTTP send-and-receive for the 8 connectors (Slack body.ok post-check, Klaviyo vnd.api+json, Brevo lowercase header, Mailchimp Basic, plus the safety gates and data substitution) was validated during development against a stdlib http.server mock. That mock harness is a dev tool and is not shipped in the repo; the shipped suite covers the resolver layer via tests/test_connector_resolver.py.
Model curator: no hardcoded model ids
Frontier models change every ~6 weeks. Hardcoding claude-sonnet-4-5-20250929 or gemini-2.0-flash across dozens of scripts means a provider deprecation silently 404s, the user blames the plugin, and the maintainer has to grep three repos. So we don’t hardcode.
scripts/model_registry.json — single source of truth for every model id used by the plugin, with vendor, tier, modality, status, and replacement_id for deprecated entries.scripts/resolve_model.py — Python module + CLI. Resolves human aliases (latest-balanced-anthropic, latest-fast-anthropic, latest-text-openai, latest-vision-google, latest-image-google, latest-video-google) to concrete ids at call time. Deprecated ids passed via --model auto-fall-forward to their replacement (with a stderr warning).scripts/refresh_models.py — polls Anthropic / OpenAI / Google / Evolink list endpoints with your API keys (every page; each key only in its provider’s request header) and reports drift versus the registry (NEW models in the provider catalog, STALE models in the registry).- Speech kinds are pins with an expiry.
latest-tts-* and latest-stt-* (OpenAI, ElevenLabs, Sarvam) can’t be discovered, because no list endpoint tells a speech model from a chat model. Each entry carries pinned_on and pin_expires, and the weekly dated check fails once a pin expires.
scripts/sync_model_registry.sh — copies the registry into the sibling plugins as a mirror: the canonical file minus the marketing-only model families. --check reports entries a copy is missing and entries upstream has dropped.scripts/model_router.py (first job: both legs of scripts/ai-visibility-checker.py) — picks the model from the provider’s own live list instead of the registry: the newest current, generally available model in the job’s line (Anthropic) or family (OpenAI, Google) from scripts/model_hints.json, at least 14 days old. A model asked for by name is kept if current, kept with a warning if deprecated, and refused if retired. One list call per provider per 7 days per key name; MODEL_ROUTER_OFFLINE=1 uses the registry aliases instead. The Google adapter has no DMP job yet; docs/MODEL-CURATOR.md explains why it ships here.
The one script that calls a model provider itself, scripts/ai-visibility-checker.py, takes --openai-model / --anthropic-model to ask for a model by name; both are routed as above. See docs/MODEL-CURATOR.md for the full alias map, curation policy, and worked examples.
python scripts/resolve_model.py --alias latest-balanced-anthropic # -> claude-sonnet-5-5
python scripts/resolve_model.py --check gpt-4o # -> deprecated (shutdown 2027-01-14; falls forward to gpt-6-luna)
python scripts/resolve_model.py --list --vendor anthropic --status current
python scripts/refresh_models.py # drift report (needs API keys)
Compliance — 16 jurisdictions, EU AI Act Article 50 ready
DM Pro carries jurisdiction-specific compliance rules that auto-apply when a brand declares its target markets. Coverage:
🇪🇺 EU (GDPR + AI Act Article 50) · 🇺🇸 US Federal (CAN-SPAM) · 🇺🇸 California (CCPA/CPRA) · US 20+ state privacy laws · 🇨🇦 Canada (CASL + PIPEDA) · 🇧🇷 Brazil (LGPD) · 🇬🇧 UK (UK GDPR + PECR) · 🇦🇺 Australia (Privacy Act + Spam Act) · 🇸🇬 Singapore (PDPA) · 🇨🇳 China (PIPL) · 🇮🇳 India (DPDPA) · 🇯🇵 Japan (APPI) · 🇰🇷 South Korea (PIPA) · 🇸🇦 Saudi Arabia (PDPL) · 🇦🇪 UAE (Federal Decree-Law No. 45) · 🇹🇭 Thailand (PDPA)
EU AI Act Article 50 (has applied since 2 August 2026; generative systems that existed before then have until 2 December 2026 to mark their output under Article 50(2)):
- C2PA content-provenance signing via
/digital-marketing-pro:c2pa-metadata (end-to-end tested against c2pa-python 0.32 — 75-byte test PNG → 42,818-byte signed PNG with manifest_embedded_and_verified=true)
- Pre-publish gate (
/digital-marketing-pro:check) treats missing C2PA on AI-flagged assets in EU campaigns as CRITICAL → BLOCKED
- Final Article 50 Guidelines + final Code of Practice on Transparency of AI-Generated Content (10 June 2026) in
compliance-rules.md §1.1b.i with six-row clarification table + five-point action list
- Production cert guide at
docs/c2pa-production-cert-guide.md covers the four CAI-recognised authorities (Adobe Content Credentials, Truepic, Numbers Protocol, Microsoft Azure Confidential Ledger)
Other regulatory updates baked in:
- NY synthetic-performer disclosure law S.8420-A (in force 2026-06-09; $1,000 for a first violation, $5,000 after) — ads that use AI-generated performers
- FTC Consumer Reviews and Testimonials Rule (in force 2024-10-21) and Endorsement Guides — fake reviews and testimonials, AI-generated ones included
- CJEU March 2026 ruling — pseudonymized cookie IDs are personal data when re-identification is feasible
- CCPA ADMT regulations (in effect 1 Jan 2026; ADMT compliance from 1 Jan 2027) + AI-derived sensitive data classification
- DPDP Phase II preparation — consent manager registration opens Nov 2026
AEO / GEO — what changed in 2026
The search landscape pivoted hard in 2025–2026:
- AI Overviews and lower click-through on the queries that show them: the figures come from vendor studies (for example Seer Interactive, September 2025) and differ by study. Google’s own statement is that clicks from AI Overviews are “higher quality”.
- Google AI Mode has more than one billion monthly users and runs on Gemini 3.5 Flash by default; AI Overviews remain the default search surface (Google, I/O 2026, 2026-05-19).
- ChatGPT: more than 1.2 billion people use it each week (OpenAI, October 2026). Which sites the engines cite (Reddit, Wikipedia) comes from vendor studies and varies by study.
DM Pro’s AEO/GEO skills (/digital-marketing-pro:aeo-audit, :geo-monitor, :entity-audit) reflect this:
- 6-platform audit standard — ChatGPT, Perplexity, Google AI Mode, Google AI Overviews, Gemini, Microsoft Copilot (was 5 — AI Mode added May 2026)
- Schema strategy — Google: "Structured d
README 内容过长，已截取前 64 KB。完整内容请前往 GitHub 仓库查看。
评论