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 mehrad-dm/mastermind --skill personagit clone --depth 1 https://github.com/mehrad-dm/mastermindWrote 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/mehrad-dm/mastermind/persona)<a href="https://agentmods.dev/skills/mehrad-dm/mastermind/persona"><img src="https://agentmods.dev/badge/skills/mehrad-dm/mastermind/persona.svg" alt="Measured on agentmods" 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 | $0.00084 | $0.01113 |
| Opus 5 | $0.00042 | $0.00557 |
| Sonnet 5 | $0.00017 | $0.00223 |
| Haiku 4.5 | $0.00008 | $0.00111 |
Grade A, and why
persona 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 — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Persona: write in the documented style of a named engineer
The user names an engineer they admire and MasterMind writes the code in that person's documented public style. Works on any code, any project, not just your own. It's a lens on taste, not a new rulebook.
What a persona actually is: the recurring, public choices that make someone's code recognizable: which primitives they reach for, how they name and factor things, their stance on abstraction vs. directness, comments, tests, error handling, and the principles they've published in talks/posts/OSS. Build the lens from real, verifiable public work: their open-source repos, their writing, their documented positions, not vibes.
Four rules, none skippable:
- Documented, not fabricated. Apply only style you can ground in their real public work, grounded
means a resolvable primary-source link in the output (see the citation gate below). If you don't
actually know how they'd write something, say so and fall back to
signature/mentors.md/ house style, never invent "they would do X" or put words/opinions in a real person's mouth. Honesty over performance. - Style is a lens; correctness is not up for grabs. A persona shapes taste: naming, structure,
idiom.
core/rigor.md, security, and a11y always outrank it. "They'd skip the null-check" is not a licence to ship a bug. If the persona and correctness conflict, correctness wins and you say why. - In the style of: not impersonation. The output is "written in the documented style of X," never a claim that X wrote it, endorsed it, or was involved. No fabricated quotes, no forged attribution, no signing their name to a commit. It's a stylistic homage, stated as one.
- Known engineers only, from public record. This is for public figures with a documented body of work.
A private colleague's style is
signature(quarantine-gated): never profile a named private individual here.
How to apply: name the 3–6 load-bearing traits of their public style, write the code through them, and in one line tell the user which traits you leaned on (e.g. "leaned on: small pure functions, minimal deps, teaching-comments, per their public work"). Prefer their principles over surface mannerisms: a persona is how someone decides, not just how they format. It composes with the active field: persona sets taste, the field pack supplies the stack, rigor supplies the gate.
Citation gate: every trait ships a resolvable link, or it doesn't ship:
- No citation, no trait. Each trait carries a URL to a primary source in the output: their repo/file/commit, their book, their talk, their own post, a style guide they authored. An actual link the user can click, not "as documented by X". A trait you cannot cite does not go in the output: dropped, not softened, not hedged, not caveated.
- Verify, don't assert. If you can reach the web in this session, actually fetch each link and confirm it resolves and says what you claim: a check you run, not one you report. If you can't, say so, cite only sources you can name precisely enough for the user to check by hand (exact repo + path, book + chapter, talk + title/year), and mark the set unverified. Never write "I checked / read / verified" for a check you didn't run.
- No sources, no persona. If you can't find primary sources for the named person at all, say so plainly
and stop, don't synthesize a plausible style from their fame or reputation. Fall back to
signature(rules from the user's actual codebase) or house style /mentors.md, and tell the user which.
A stereotype ("they'd use lots of patterns") is the failure mode; the cited habit is the goal.
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 First seen · 62 lines · 0 tokens per session scan A 762d0701ea4e
persona is a skill published in the GitHub repository mehrad-dm/mastermind (24 stars, last pushed 8d ago), licensed MIT. It adds 84 tokens to every session and 1,113 once invoked, about $0.0004 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
writing-git-commits
Write clear, focused Git commit messages and keep commits easy to review.
st-full-workflow
Use when the user asks to run the complete end-to-end Strikethroo workflow for a work order in one shot in this repository — triggers include full workflow, end-to-end, plan and execute, do everything, run the whole strikethroo workflow. Do not use when the user wants only one stage (create a plan, generate tasks, or…
st-refine-plan
Use when the user asks to review, refine, improve, interrogate, pressure-test, or update an existing Strikethroo plan by plan ID in this repository — triggers include refine plan, improve plan, review plan, red-team the plan, update plan. Do not use to create a new plan, to generate tasks, or for generic brainstorming…
hatch3r-board-refresh
Regenerates the living board overview dashboard from current board state. Scans all open issues, computes health metrics, and updates the meta:board-overview issue.
hatch3r-qa-path
Produces a human-run manual QA test path from a PR, branch diff, or uncommitted working tree — a risk-ordered table of steps, expected results, and automated-coverage references plus a shippability sign-off. Rows are auto-proven via delegated verification sub-agents before emission, so the human walks only what…
hatch3r-recipe
Authors and validates composition specs that an orchestrating agent walks via the Task tool to run hatch3r commands and skills in a dependency-ordered sequence. Use when designing a multi-step capability composition, customizing an existing one, or debugging a composition the agent walks.