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 agentmods add skills/mehrad-dm/mastermind/prototypenpx skills add mehrad-dm/mastermind --skill prototypegit clone --depth 1 https://github.com/mehrad-dm/mastermindWhat 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 | $0.00050 | $0.00562 |
| Opus 5 | $0.00025 | $0.00281 |
| Sonnet 5 | $0.00010 | $0.00112 |
| Haiku 4.5 | $0.00005 | $0.00056 |
Grade A, and why
prototype 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 2d 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 — 41 lines — stays where its author put it; the contents beside it link to each section on GitHub.
MasterMind: Prototype
Tight feedback loops beat big plans when the path is unclear (~/.mastermind/engineering/core/agent-loop.md, mindset.md →
design it twice). A prototype buys knowledge cheaply: the discipline is keeping it knowledge, not product.
Method
- Name the question: the ONE risky unknown this prototype will answer ("can the Worker stream a >20 MB file?", "does this library handle X?"). One named question is the entry price, find it first.
- Build the smallest thing that answers it: hard-code, skip error handling, skip tests, skip polish. Speed over quality; it's throwaway.
- Extract the learning: what did reality teach? Write down the answer and any surprises.
- Throw it away and rebuild to standard: the real version gets the full loop (design, rigor, tests, review). Ship the rebuild; the prototype gets discarded.
Rules
Time-box: 5 build-and-run attempts, roughly 30 minutes: one question needs a handful of tries, not a day; proportionality over ceremony. An attempt is one edit → run → observe cycle against the named question; count them out loud as you go, so the box is observable and not a vibe. Attempt 5 finished without an answer = the box is spent, whatever the clock says.
When it expires, stop. That's a result, not a failure. Report: what you learned, what is still unknown, and one recommendation: proceed (answer is yes, build it properly) / different approach (this path is wrong, here's the next one) / needs more investigation (name what a second spike would target). Then discard the code: it stays throwaway and does not ship, same as a prototype that succeeded.
Extending the box is a decision, not drift. Only extend when the attempts narrowed the question rather than wandering: say so explicitly, state the new bound (e.g. "3 more attempts"), and stop there.
Keep it isolated (a scratch branch/dir) so it can't leak into production. The deliverable is the learning, not the code.
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.
- 2d ago First seen · 41 lines · 0 tokens per session scan A 8d9a4be83fa8
prototype is a skill published in the GitHub repository mehrad-dm/mastermind (24 stars, last pushed 2d ago), licensed MIT. It adds 50 tokens to every session and 562 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 skills, from other repositories
skill-creator
Create, edit, validate, and package new Tlamatini skills (SKILL.md packages).
gstack-roles
Role-based AI specialist system. Defines 23 specialist roles and teaches the AI to adopt the correct role before each task phase, like a virtual engineering team.
learning-memory
Use when capturing or restoring a learner's persistent profile to personalize teaching across sessions.
your-skill-name
State the educational problem this skill solves and the intended outcome.
interview-mode
Use when simulating a technical interview to build confidence, expose reasoning gaps, and train structured communication under pressure.
misconception-detector
Use when diagnosing a repeated conceptual mistake and designing a targeted correction loop to replace the faulty mental model.