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.
npx skills add iamwaqargulzar/Marketing-Agent-OS --skill conversion-trackinggit clone --depth 1 https://github.com/iamwaqargulzar/Marketing-Agent-OSWrote 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/skills/iamwaqargulzar/marketing-agent-os/conversion-tracking)<a href="https://agentmods.dev/skills/iamwaqargulzar/marketing-agent-os/conversion-tracking"><img src="https://agentmods.dev/badge/skills/iamwaqargulzar/marketing-agent-os/conversion-tracking/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/skills/iamwaqargulzar/marketing-agent-os/conversion-tracking"><img src="https://agentmods.dev/badge/skills/iamwaqargulzar/marketing-agent-os/conversion-tracking.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 | $0.00080 | $0.00920 |
| Opus 5.5 | $0.00032 | $0.00368 |
| Sonnet 5.5 | $0.00016 | $0.00184 |
| Haiku 4.5 | $0.00008 | $0.00092 |
Grade A, and why
conversion-tracking 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 14d 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 — 44 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Conversion Tracking
Measure the defined business outcome and verify its delivery, rather than equating a tag firing with a received conversion. Read references/skill-contract.md for evidence, permissions, and handoffs.
Scope and permissions
Start with the exact user-named site/property, desired outcome, destinations, existing implementation, and available evidence. Read relevant product context when present; do not invent missing setup or explore unrelated connected accounts.
A diagnosis request authorizes investigation, not implementation. A live form submission, booking, order, test conversion, tag publication, account edit, or upload can change records and bidding signals. Use an authorized staging/sandbox fixture where possible; obtain explicit scope for real submissions and writes. Report a blocked validation step instead of manufacturing a receipt.
Setup or audit
- Define the conversion precisely: accepted lead, confirmed booking, account created, paid order, qualified opportunity, or another specified outcome. A phone-link click measures click intent, not a completed call; a file click does not prove a completed download.
- Inventory existing browser tags, native platform integrations, backend events, CRM jobs, and offline uploads. Browser inspection cannot establish that no server-side sender exists. Ask for the unresolved sender inventory before adding another source.
- Locate the authoritative success signal. Prefer the backend/order/accepted-submission record when available. Otherwise use the provider's documented success callback/event for the installed version. Do not substitute submit clicks, popup close, page exit, or an unguarded thank-you-page visit.
- Check existing consent handling, destination eligibility, field provenance, event/value/currency mapping, identifiers, deduplication, and error/retry behavior. Do not put raw contact details in URLs, analytics events, console output, or generic dataLayer payloads. Hashing does not remove permission and data-use requirements.
- Choose a delivery option from
references/delivery-options.md; preserve a working native path when adequate. A server-side sender does not bypass consent, access restrictions, or a missing success signal. - For diagnosis, report the supported cause, excluded causes, unknowns, and smallest reversible proposed change. For an explicitly requested fix/setup, implement only the authorized change and retain a rollback path.
- Follow
references/verification.mdto prove detection, sending, destination receipt, and deduplication separately. If account access, a safe test, or sufficient observation time is unavailable, name the exact unverified boundary.
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 14d ago First seen · 44 lines · 80 tokens per session scan A b039d579db67
conversion-tracking is a skill published in the GitHub repository iamwaqargulzar/Marketing-Agent-OS (8 stars, last pushed 3d ago), licensed Apache-2.0. It adds 80 tokens to every session and 920 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-10-23.
conversion-tracking compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| conversion-trackingiamwaqargulzar/Marketing-Agent-OS | Skill | 8 | 3d ago | Apache-2.0 | A | 80 | Claude Code |
| ansvisor-aeo-coach-standaloneansvisor/ansvisor | Skill | 232 | yesterday | MIT | A | 148 | Claude Code, Codex |
| aeroCanonry/canonry | Skill | 198 | yesterday | MIT | A | 93 | Claude Code |
| refreshansvisor/ansvisor | Skill | 232 | yesterday | MIT | A | 17 | Claude Code, Codex |
| wrongclaimansvisor/ansvisor | Skill | 232 | yesterday | MIT | A | 17 | Claude Code, Codex |
| seo-vertical-detectHainrixz/claude-seo-ai | Skill | 82 | 13d ago | MIT | A | 71 | Claude Code |
| mcp-geoAKzar1el/mcp-geo | Skill | 40 | yesterday | MIT | A | 42 | Claude Code |