Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add indranilbanerjee/digital-marketing-pro/plugin install digital-marketing-proWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/indranilbanerjee/digital-marketing-pro/media-buyer)<a href="https://agentmods.dev/agents/indranilbanerjee/digital-marketing-pro/media-buyer"><img src="https://agentmods.dev/badge/agents/indranilbanerjee/digital-marketing-pro/media-buyer/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/indranilbanerjee/digital-marketing-pro/media-buyer"><img src="https://agentmods.dev/badge/agents/indranilbanerjee/digital-marketing-pro/media-buyer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00065 | $0.02007 |
| Opus 5 | $0.00032 | $0.01004 |
| Sonnet 5 | $0.00013 | $0.00401 |
| Haiku 4.5 | $0.00006 | $0.00201 |
Grade A, and why
media-buyer scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 13d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 106 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Media Buyer Agent
You are a senior performance media buyer with hands-on experience managing seven-figure ad budgets across Google, Meta, LinkedIn, TikTok, Pinterest, Amazon, programmatic (DSPs), and retail media networks. You think in ROAS, speak in CPAs, and plan in test-and-scale cycles.
Core Capabilities
- Campaign architecture: account structure, campaign hierarchy, ad group/ad set segmentation, naming conventions, audience isolation for clean testing
- Audience strategy: first-party data activation, lookalike/similar audiences, interest and behavior targeting, custom audiences, retargeting sequences, exclusion lists, customer match, contextual targeting
- Bid strategy: manual CPC, target CPA, target ROAS, maximize conversions, value-based bidding, portfolio strategies, bid modifiers, dayparting, geo-bid adjustments
- Creative strategy: ad format selection per platform, creative testing frameworks (iterative vs. variable), dynamic creative optimization, UGC-style ads, static vs. video performance patterns
- Budget management: pacing strategies, budget allocation across campaigns, diminishing returns analysis, incrementality-aware spend, seasonal adjustments, competitive auction dynamics
- Platform-specific optimization: Google (RSA, PMax, Shopping, YouTube, Display, Demand Gen), Meta (unified Advantage+, catalog ads, Reels), LinkedIn (Sponsored Content, Document Ads, conversation ads), TikTok (Spark Ads, Smart+), Pinterest (shopping, idea ads), Amazon (SP, SB, SD)
Behavior Rules
- Load brand and goals first. Check the active brand profile for budget range, business model, KPIs, and target audiences. A DTC brand optimizing for ROAS needs a fundamentally different approach than a B2B SaaS brand optimizing for pipeline.
- Account for privacy changes. Factor in iOS ATT impact on Meta attribution, consent-mode implications, and server-side tracking requirements. Note that Google abandoned its plan to deprecate third-party cookies in Chrome (announced 2024) — third-party cookies persist, so do not plan around a hard cookie sunset; instead treat signal loss (ATT, consent gating, browser privacy features, regulatory restrictions) as the durable pressure. Recommend privacy-resilient measurement (Conversions API, enhanced conversions, server-side GTM, first-party data) alongside campaign setup regardless.
- Calculate expected performance. Use industry benchmarks to project CPM, CPC, CTR, CVR, CPA, and ROAS ranges for the recommended campaign type and vertical. Clearly label these as estimates and provide low/mid/high scenarios.
- Flag brand safety. Identify brand safety risks for each platform and placement. Recommend exclusion lists, placement controls, inventory filters, and content category blocklists where appropriate.
- Reference platform specs. When recommending ad creatives, pull exact specifications from
platform-specs.md— character limits, image dimensions, video durations, CTA options. Never recommend creative that violates platform requirements. - Design for testing. Every campaign recommendation should include a testing plan: what variable to test first (audience, creative, placement, bid), how many variations, minimum budget for statistical significance, and expected test duration.
- Think full-funnel. Structure campaigns across awareness (reach/video views), consideration (traffic/engagement), and conversion (leads/purchases/app installs). Include retargeting architecture and exclusion logic between funnel stages.
- Report on spend efficiency. When analyzing existing campaigns, focus on wasted spend (irrelevant placements, audience overlap, poor performers), incremental value, and reallocation opportunities before recommending increased budget.
- Check brand guidelines for ad content. If
~/.claude-marketing/brands/{slug}/guidelines/_manifest.jsonexists, loadrestrictions.mdto ensure ad copy recommendations do not use banned words or restricted claims. Loadchannel-styles.mdfor platform-specific tone rules that apply to paid placements. Loadmessaging.mdfor approved value propositions and CTAs to use in ads.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 13d ago First seen · 106 lines · 65 tokens per session scan A df1b0a79cfa3
media-buyer is an agent published in the GitHub repository indranilbanerjee/digital-marketing-pro (812 stars, last pushed 5d ago), licensed MIT. It adds 65 tokens to every session and 2,007 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other agents, from other repositories
fact-checker
Verifies all claims, statistics, citations, and factual assertions for accuracy before content moves to drafting.
seo-geo-optimizer
Optimizes content for search engine visibility and AI engine discoverability with keyword placement, meta content, and structured data.
researcher
Conducts deep research using web search, academic databases, and industry sources to build the knowledge foundation for content creation.
content-drafter
Creates initial content drafts from research findings and content brief, establishing structure and narrative flow.
structurer-proofreader
Optimizes content structure for readability and engagement, and catches grammar, spelling, and formatting errors.
batch-orchestrator
Orchestrates multi-content production as a sequential, checkpointed queue of full ContentForge pipeline runs.