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 meikocho1/alchemy-marketing-skills --skill costly-signalinggit clone --depth 1 https://github.com/meikocho1/alchemy-marketing-skillsWrote 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/meikocho1/alchemy-marketing-skills/costly-signaling)<a href="https://agentmods.dev/skills/meikocho1/alchemy-marketing-skills/costly-signaling"><img src="https://agentmods.dev/badge/skills/meikocho1/alchemy-marketing-skills/costly-signaling/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/meikocho1/alchemy-marketing-skills/costly-signaling"><img src="https://agentmods.dev/badge/skills/meikocho1/alchemy-marketing-skills/costly-signaling.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 | $0.00224 | $0.01504 |
| Opus 5.5 | $0.00090 | $0.00602 |
| Sonnet 5.5 | $0.00045 | $0.00301 |
| Haiku 4.5 | $0.00022 | $0.00150 |
Grade A, and why
costly-signaling 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 2mo 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.
This is a copy
86% identical to local-mirror — 355 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 58 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Costly Signaling — trust is bought with waste
People can't see your intentions, only your actions — so they read your costly, hard-to-fake actions as proof. Anything cheap and easy to fake (a slogan, a claim, a promise) carries little signal. Anything expensive, effortful, or risky to you carries a lot. This is why "wasteful" things work.
The principle
A signal is credible in proportion to how much it would cost a liar. The peacock's tail is useless and metabolically expensive — that's exactly why it's a believable signal of fitness. Apply the same logic to a brand:
- Advertising spend is itself the message. A big, visible campaign says "we expect to be here long enough to recoup this, and we have repeat customers." A fly-by-night operator can't afford it. The customer doesn't need to read the ad to get the signal.
- Guarantees, warranties, free returns are costly signals: you'd only offer them if the product actually held up. The cost to you is the proof.
- Over-delivery and effort ("we'll answer in three rings," lavish packaging, a hand-written note) signal commitment because they obviously cost something.
- Price can be the feature. For Veblen goods (champagne, perfume, premium tools), a higher price increases desirability and even perceived performance — the expensive placebo genuinely works better. Cutting the price can destroy the signal.
The Pratfall Effect (the honest-flaw signal)
Admitting a real weakness is a costly, hard-to-fake move — so it reads as honesty, and honesty raises trust in everything else you say.
- Avis: "We're number two. We try harder." Owning the weakness made the brand more likeable and more credible than claiming to be the best.
- "It's ugly, but it works." / "We're not the cheapest, but here's why." A small genuine flaw makes the rest believable. Perfection reads as marketing; an admitted flaw reads as a person.
How to apply
- Read your current signal. Forget what you say — what do your actions cost you, and therefore prove? Cheap claims ("trusted," "premium," "#1") prove nothing.
- Find a signal a liar couldn't afford. A guarantee that would bankrupt you if the product were bad. Visible, sustained presence. Effort the customer can see. Real expertise on display.
- Consider price as a signal, not just a cost. If trust/desire is the problem, a higher price plus a costlier-looking presentation can outperform a discount. Test before assuming cheaper wins.
- Add one honest flaw (Pratfall). State a real limitation plainly. It buys credibility for every true claim around it. Never invent a fake flaw — that's just another cheap claim.
- Make sure the signal is honest. A costly signal works because it's tied to reality (you really will honour the guarantee, you really are committed). Faking it is expensive to you and fatal when discovered.
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.
- 2mo ago First seen · 58 lines · 224 tokens per session scan A 794cf88cd484
costly-signaling is a skill published in the GitHub repository meikocho1/alchemy-marketing-skills (4 stars, last pushed 1mo ago), licensed MIT. It adds 224 tokens to every session and 1,504 once invoked, about $0.0009 per session on Opus 5.5. A static security scan graded it A with 0 findings. It is 86% identical to local-mirror, differing in 355 lines, and is treated as a copy.
costly-signaling compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| costly-signalingmeikocho1/alchemy-marketing-skills | Skill | 4 | 1mo ago | MIT | A | 224 | Claude Code |
| ads-performance-analyticsrampstackco/claude-skills | Skill | 1.0k | 11d ago | MIT | A | 147 | Claude Code |
| paid-media-strategyrampstackco/claude-skills | Skill | 1.0k | 11d ago | MIT | A | 142 | Claude Code |
| ecommerce-full-pipelineanbeime/skill | Skill | 9.2k | yesterday | none | A | 96 | Claude Code, Codex |
| yandex-direct-managerai-hub-open/yandex-direct-manager | Skill | 14 | 8d ago | Apache-2.0 | A | 214 | Claude Code |
| marketing-strategistai-hub-open/marketing-strategist | Skill | 12 | 8d ago | Apache-2.0 | A | 66 | Claude Code |
| vk-ads-managerai-hub-open/vk-ads-manager | Skill | 3 | 8d ago | Apache-2.0 | A | 324 | Claude Code |