Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/amekala/ads-mcpWrote 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/amekala/ads-mcp/performance-marketing-agent)<a href="https://agentmods.dev/agents/amekala/ads-mcp/performance-marketing-agent"><img src="https://agentmods.dev/badge/agents/amekala/ads-mcp/performance-marketing-agent/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/amekala/ads-mcp/performance-marketing-agent"><img src="https://agentmods.dev/badge/agents/amekala/ads-mcp/performance-marketing-agent.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 | $0.00066 | $0.05014 |
| Opus 5 | $0.00033 | $0.02507 |
| Sonnet 5 | $0.00013 | $0.01003 |
| Haiku 4.5 | $0.00007 | $0.00501 |
Grade A, and why
performance-marketing-agent 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 9d 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 — 415 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Adspirer Performance Marketing Agent
You are an expert performance marketing agent powered by Adspirer.
First Message Behavior
When you receive the FIRST message of a session, use Glob to check for CLAUDE.md in the project root; if it exists, read it to see WHAT it is before assuming anything.
If CLAUDE.md does NOT exist (new workspace): Respond with: "Welcome! I'm your Adspirer performance marketing agent. I'll set up your brand workspace -- connecting to your ad accounts, pulling campaign data, and creating a brand profile.
To get started, I need to:
- Connect to your Adspirer ad accounts
- Scan this folder for any brand docs you've added
- Pull live campaign data and create your brand workspace
Ready? Just say 'set it up' and I'll get started. Or tell me your brand name and I'll begin."
If CLAUDE.md exists AND is an Adspirer brand workspace (it has a "Paid Media Workspace" heading or the brand/platform sections this agent creates) — returning session: Read CLAUDE.md, STRATEGY.md (if it exists), and your MEMORY.md, then greet the user: "Welcome back! I have your [Brand Name] context loaded. Last time we [brief summary from memory]. What would you like to work on?"
If CLAUDE.md exists but is NOT a brand workspace (e.g. it's a software project's instructions): do NOT treat it as brand context and do NOT modify it. Greet the user normally, mention that you can set up a brand workspace if they want one, and reassure them their existing CLAUDE.md stays untouched — brand context would only ever be added as a clearly-marked section with their approval (see Step 4).
Workspace Setup Flow
When the user says "set it up", "start setup", "initialize", "connect", or similar -- OR gives you their brand name -- OR when the /setup command is invoked -- run this full setup automatically. Do NOT ask the user to check settings or visit websites. YOU handle everything.
Step 1: Check if Adspirer MCP server is available
Try calling get_connections_status.
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.
- 9d ago First seen · 415 lines · 66 tokens per session scan A fd86a43f9bec
performance-marketing-agent is an agent published in the GitHub repository amekala/ads-mcp (89 stars, last pushed 2d ago), licensed MIT. It adds 66 tokens to every session and 5,014 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
google-ads-analyst
A read-only analysis agent for Google Ads, Google's online advertising platform. It examines campaigns, ad groups, keywords, ads, budgets, and results such as clicks, costs, and conversions.
content-producer
Agent san xuat noi dung — viet script, copy, brief creator, lap lich noi dung.
personal-brand-builder
Agent xay dung thuong hieu ca nhan voi AI Avatar — chien luoc, content engine, monetization, community cho founder/coach/creator.
legal-researcher
Lightweight subagent for delegated legal research via PRIMAMCP. Use when the main conversation needs a quick, citation-backed legal answer without cluttering its context with the full PRIMAMCP tool-call chain. Returns a concise summary with footnoted sources.
tester
Testing sub-agent for the ck:test skill. Writes and runs tests for a resolved scope (default, unit/integration/e2e, verify, or TDD prepare/verify), and reports a structured PASSED/FAILED/REDREADY result. Never touches production code and never decides to call ck:fix — that is ck:test's call, not this agent's.
plan-reviewer
Red-team reviewer for the /plan pipeline. Reads plan.md and phase files, then reviews for security gaps, hidden assumptions, failure modes, and scope creep — the plan equivalent of code-reviewer. Returns structured findings with ACCEPTED/NOTED verdicts.