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 stefanoskarakasis/Product-Marketing-Skills --skill experiment-ideasgit clone --depth 1 https://github.com/stefanoskarakasis/Product-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/stefanoskarakasis/product-marketing-skills/experiment-ideas)<a href="https://agentmods.dev/skills/stefanoskarakasis/product-marketing-skills/experiment-ideas"><img src="https://agentmods.dev/badge/skills/stefanoskarakasis/product-marketing-skills/experiment-ideas/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/stefanoskarakasis/product-marketing-skills/experiment-ideas"><img src="https://agentmods.dev/badge/skills/stefanoskarakasis/product-marketing-skills/experiment-ideas.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 | $0.00083 | $0.02499 |
| Opus 5.5 | $0.00033 | $0.01000 |
| Sonnet 5.5 | $0.00017 | $0.00500 |
| Haiku 4.5 | $0.00008 | $0.00250 |
Grade A, and why
experiment-ideas 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 8d 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 — 268 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Experiment Ideas
How This Works
Most idea-brainstorm output is five generic tactics that could belong to
any company — "post more on LinkedIn," "run a referral program." This
skill grounds ideation in what's actually known: who the ICP is, what
alternatives buyers compare against, and why now matters, so every idea
is specific enough to differentiate from what a status-quo competitor is
already doing. It doesn't write anything durable — generated ideas are
disposable brainstorm output, not a company fact — and it hands off
explicitly to experiment-doc (pmm-execution), which is where a chosen
idea gets pressure-tested before it becomes a real test.
Step 0 — Load brain Sections 1 (Product), 2 (ICP), 3 (Alternatives),
5 (Market Context) if present, and guardrails from
/context/meta-patterns.md.
Step 1 — Clarify constraints: budget or resource ceiling, channels already tried (and what happened), how many ideas wanted (default 5).
Step 2 — Generate ideas: each with Channel, Core Message, Why It Works (tied to ICP or market energy state when brain data exists), Cost Efficiency.
Step 3 — Self-check each idea against the ICP and named alternatives — an idea indistinguishable from what a status-quo competitor already does gets flagged, not silently kept.
Step 4 — Rank by effort-vs-impact, not generation order.
Step 5 — Learning Close: log the session to /context/skill-sessions.md.
Trigger
-
When: Brainstorming marketing or growth ideas, looking for low-budget/high-impact tactics, exploring channels before committing budget, or any "what should we try" moment before a specific hypothesis exists.
-
Not for: Pressure-testing a single idea you've already picked →
experiment-doc(pmm-execution), run this skill first only if you don't yet have a specific hypothesis to test. Rewriting an already-set positioning into segment-specific value-prop copy →value-prop-statements(pmm-growth), run this skill first only if you need raw campaign ideas, not copy variants of an existing statement. A full campaign brief with calendar, audience, and metrics →gaccs-brief. Channel or motion selection for a launch already scoped against ICP deal economics →gtm-motions. Writing the actual positioning statement or messaging hierarchy →positioning-messaging.
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.
- 8d ago Changed · -1 lines 4253d7501c45
- 16d ago Changed fd5b40b269b2
- 17d ago Changed 2e4b80288ed0
- 20d ago Changed · +1 lines 6ff4e7c57bc0
- 21d ago Changed · +2 lines 68e0c98cdd88
- 22d ago Changed · +12 lines 19524261d30d
- 1mo ago Changed · +3 lines 39cc99f8e116
- 1mo ago First seen · 251 lines · 83 tokens per session scan A 7b7a382ab493
experiment-ideas is a skill published in the GitHub repository stefanoskarakasis/Product-Marketing-Skills (8 stars, last pushed yesterday), licensed MIT. It adds 83 tokens to every session and 2,499 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 2026-09-04.
Other skills, from other repositories
report
A way to export a chat card or reply as a standalone HTML report, meaning a file that can open in a browser without the chat application. It preserves the card template, including right-to-left Persian or left-to-right English layouts.
k8s-rightsize
Right-size Kubernetes CPU and memory requests from Prometheus usage (p95 CPU, p98 memory over 7 days), size the node count, tune KEDA autoscaling. Use when asked to rightsize workloads, cut node cost, or "how many nodes do we need".
owasp-security
Deep OWASP reference: Top 10 2025, ASVS 5.0, secure patterns, per-language quirks, LLM Top 10 and agentic AI security. Use for an in-depth security review, auth work, or hardening an LLM or agent app. For the everyday baseline use secure-coding.
fig
One self-contained HTML file holding a looping animated SVG, for ideas that move: flows, loops, retries, queues, fan-outs. Use when the user says "fig" or a diagram would be better animated.
brand
A project-specific visual identity for chat cards and saved reports, using the project's colors, logo, and fonts. It finds these details in the repository or asks the user for them, then stores the result in a committable `.readable/` folder.
workspace-rules
A set of guidelines for deciding which rules belong in a repository's workspace instructions and which should be enforced by its build or checks.