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 double-checkgit 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/double-check)<a href="https://agentmods.dev/skills/mehrad-dm/mastermind/double-check"><img src="https://agentmods.dev/badge/skills/mehrad-dm/mastermind/double-check/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/mehrad-dm/mastermind/double-check"><img src="https://agentmods.dev/badge/skills/mehrad-dm/mastermind/double-check.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 2 findings, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Rogue Agent · line 18 Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.Fix: Remove any persistence mechanisms (cron jobs, startup scripts, state files). Skills should not maintain state across sessions without explicit user consent.
- medium Agent Snooping · line 57 Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.Fix: Remove all code or instructions that list or read other skills' files or directories. Skills should operate independently; cross-skill access is a privilege escalation.
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 | $0.00129 | $0.01229 |
| Opus 5.5 | $0.00052 | $0.00492 |
| Sonnet 5.5 | $0.00026 | $0.00246 |
| Haiku 4.5 | $0.00013 | $0.00123 |
Grade A, and why
double-check 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 15d 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 — 80 lines — stays where its author put it; the contents beside it link to each section on GitHub.
MasterMind: Double-check
You are about to hand over something you believe. Belief is not evidence
(~/.mastermind/engineering/core/rigor.md → Report against evidence), and the check that catches a wrong
claim has to happen while the work is still moving, not after the user has acted on it. Claim:
$ARGUMENTS.
The kernel's rule: whatever did the work doesn't get to grade it: is the entire mechanism here. This applies it to a claim, not only to a diff.
The five moves
-
Claim. Write the belief you are about to act on or hand over, in one sentence: "the retry loop now handles the expired-token case." A paragraph means several claims, doubt them one at a time. The claim stays with you; it is the thing under test, not context to share.
-
Extract. Split it into the artifact (the code, the design, the answer, the diagnosis) and the contract it must satisfy (the requirement, the spec line, the invariant, the question actually asked). If you can't write the contract down, stop and get one. Without it there is no bar, and any reviewer defaults to grading your intent.
-
Doubt. Dispatch a reviewer in a fresh context with the artifact and the contract, nothing else. It is a fresh context, not a second model: say so, and use a second model where one is installed. Brief it adversarially:
Find what is wrong with this. Assume the author is overconfident. Do not validate and do not summarize: report what fails the contract, or state plainly that you could not find anything after thorough examination.
Hand the artifact over via a file or stdin. Never interpolate it into a shell-quoted argument: code and prompts carry backticks,
$(...), and quotes that truncate the message or execute in your shell. A clipped artifact reviews clean for the worst reason: the flaw didn't survive the paste.Never pass it your claim. The excuse is always "it'll review faster with the context": but the context you're about to add is your conclusion, and a reviewer handed a conclusion hunts for reasons it holds. You would be buying agreement and calling it review.
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.
- 15d ago Changed · +1 lines · +129 tokens per session d5b08256a5e0
- 1mo ago First seen · 79 lines · 0 tokens per session scan A 0633b61b2f02
double-check is a skill published in the GitHub repository mehrad-dm/mastermind (23 stars, last pushed 6d ago), licensed MIT. It adds 129 tokens to every session and 1,229 once invoked, about $0.0005 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
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…
code-sec-review
Static security review that hunts exploitable vulnerabilities, prioritizing unauthenticated and cross-tenant attack paths.
agent-decision
Deterministic self-evaluation rubric for decision escalations — scored every run using the FRAME framework.
code-quality-review
Reviews code and plans against the project's coding rules.
review-loop
Two-mode review loop — single reviewer per epic, three reviewers for full branch.
setup-new-acpx-key
Configure the API key (or other credential) for a registered ACPX agentid (claude/codex/cursor/gemini/qwen/copilot/pi/droid/iflow/kilocode/kimi/kiro/opencode/tlamatini) end-to-end across data.keys, config.json (top-level + acpx.agents. .env), regensecrets.py, and verification via acpdoctor.