headcount is an organization of independently installable Claude Code plugins, each grouping skills for a department such as finance, security, or demand generation. Claude Code users install the departments they need and invoke their skills for specialized work; the catalogue entries are skills and related agent tooling from that organization.
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 cbrock84/headcount --skill experimentationgit clone --depth 1 https://github.com/cbrock84/headcountWrote 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/cbrock84/headcount/experimentation)<a href="https://agentmods.dev/skills/cbrock84/headcount/experimentation"><img src="https://agentmods.dev/badge/skills/cbrock84/headcount/experimentation/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/cbrock84/headcount/experimentation"><img src="https://agentmods.dev/badge/skills/cbrock84/headcount/experimentation.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00068 | $0.00834 |
| Opus 5.5 | $0.00027 | $0.00334 |
| Sonnet 5.5 | $0.00014 | $0.00167 |
| Haiku 4.5 | $0.00007 | $0.00083 |
Grade A, and why
experimentation 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 1mo 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 — 76 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Experimentation
Most A/B testing programs produce confident conclusions from insufficient data. The discipline is almost entirely in what you do before launch.
Before running
- Hypothesis with a mechanism. "Moving the pricing table above the fold will raise trial starts, because visitors currently leave before seeing pricing." Not "let's try a green button."
- One primary metric, chosen in advance. Secondary metrics are context, never the verdict.
- Sample size calculated in advance, from your baseline rate and the smallest lift that would change a decision. If the required sample is unreachable, do not run the test — decide by judgment and say so.
- Duration set in advance, covering at least one full weekly cycle, and two if the buying cycle is long.
- Guardrail metrics that would make you reject a win: refunds, support volume, downstream retention.
While running
Do not look at results and act on them mid-flight. Peeking and stopping at significance is the single most common way to generate false positives, and it is very effective at it.
Check only that the test is running correctly — even split, no broken variant, tracking firing.
Reading
- At the pre-set duration, not before, and not extended because it is nearly significant. Extending until significance manufactures it.
- Significance is not size. A statistically significant 0.3% lift may not be worth shipping.
- Inconclusive is a real result and the most common one. It means the change did not matter enough to detect, which is useful.
- Check the guardrails before declaring a win.
- Segment afterward for hypotheses only, never for verdicts. Slice enough ways and something is always significant.
Program level
Test where the traffic and the leverage are. Most sites can only run a handful of adequately powered tests a year — spend them on structural questions, not button colors.
Keep a log of every test: hypothesis, result, decision. Without it, teams re-run the same tests every eighteen months and re-learn the same things.
What ships with it
1 file 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.
- 1mo ago Changed · +7 lines 1112191cd4bb
- 1mo ago Changed · +20 lines b62b173f3490
- 1mo ago First seen · 49 lines · 68 tokens per session scan A dbf3251dc3d6
experimentation is a skill published in the GitHub repository cbrock84/headcount (2,069 stars, last pushed yesterday), licensed MIT. It adds 68 tokens to every session and 834 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 2026-08-30.
Other skills, from other repositories
negative-keyword-sweep
Find wasted spend in Google Ads search terms and add negatives at the right scope (account list, campaign, or ad group). Use when the user says "find junk keywords", "stop wasting on bad searches", or wants a search-terms cleanup.
rotate-creative
Pause fatigued ads and ship fresh creative variants in their place. Works on both Meta and Google Ads. Use when CTR is dropping, frequency is climbing, or the user says "creative is tired" / "let's refresh ads". Always starts new variants paused.
launch-google-search
Launch a complete Google Ads Search campaign — campaign, ad group, responsive search ads, keywords, sitelinks, callouts, and negatives — built paused and ready to review. Use when the user wants to spin up a new Search campaign on Google Ads.
go
Compatibility entry for Visual Stack's former /vstack:go command. Runs the wireframe and UI review tool, now called review. Use only when the user invokes /vstack:go.
save
Save durable decisions, findings, plans, and implementation knowledge from the visible Codex conversation into a new or existing NeatContext context. Use only when the user explicitly invokes this skill or asks to preserve the current conversation as reusable context.
dns
This skill should be used when the user asks about "DNS", "domain resolution", "GeoDNS / geo routing", "latency-based routing", "weighted / failover routing", "Route 53 / Cloud DNS", an "A/CNAME/ALIAS record", "anycast", or "DNS TTL / propagation". It gives the global front door that maps a name to the right IP and…