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 debuggit 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/debug)<a href="https://agentmods.dev/skills/mehrad-dm/mastermind/debug"><img src="https://agentmods.dev/badge/skills/mehrad-dm/mastermind/debug/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/debug"><img src="https://agentmods.dev/badge/skills/mehrad-dm/mastermind/debug.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, 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 Excessive Agency · line 70 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00063 | $0.01502 |
| Opus 5.5 | $0.00025 | $0.00601 |
| Sonnet 5.5 | $0.00013 | $0.00300 |
| Haiku 4.5 | $0.00006 | $0.00150 |
Grade A, and why
debug scanned grade A with 1 finding 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
cheapest rung that gives you that: a failing test · a `curl` · a CLI invocation with a saved How it starts
The opening of the file, as written. The whole thing — 91 lines — stays where its author put it; the contents beside it link to each section on GitHub.
MasterMind: Debug
Hard bugs are lost by guessing and patching symptoms. This is the disciplined loop that finds the
actual cause. Bug: $ARGUMENTS. Grounded in ~/.mastermind/engineering/core/rigor.md and ~/.mastermind/engineering/core/agent-loop.md.
The cardinal rule: understand the mechanism before you change a line. A fix you can't explain is a coincidence, not a fix.
The six phases
-
Build the feedback loop. This phase is the skill: the rest is mechanical once it exists.
You need one command you have already run at least once, whose invocation and output you can paste. It has to be red-capable (it fails right now, because of this bug), deterministic, fast enough to run repeatedly, and runnable by you without a human driving a UI. Reach for the cheapest rung that gives you that: a failing test · a
curl· a CLI invocation with a saved snapshot · a headless browser script · a replayed captured request · a throwaway harness · a property/fuzz loop · a bisection harness · a differential run against a known-good version.No red-capable command, no phase 2. If you catch yourself reading code to build a theory before that command exists, stop and go build it, theorising over a bug you cannot trigger is the single most expensive habit in debugging. This same command is your definition of "fixed." If it still won't reproduce after honest effort: stop: a bug you can't reproduce earns a fix only once it does, and never claim it's fixed. Ship instrumentation instead (logging/assertions at the suspect boundaries) so the next occurrence is diagnosable, state explicitly what you ruled out and how, and hand back with that evidence. "I couldn't reproduce it" is the honest answer, and usually a faster route to the real cause than three speculative patches.
-
Localize. Bisect the space: git bisect across commits, binary-search the code path, add instrumentation/logs at boundaries. Read the actual code and the actual data/state at the failure point rather than theorizing from memory. Narrow to the smallest region that still reproduces.
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 · +2 lines · +63 tokens per session d60a63ba4b3e
- 1mo ago First seen · 89 lines · 0 tokens per session scan A 9576bbaec725
debug is a skill published in the GitHub repository mehrad-dm/mastermind (23 stars, last pushed 6d ago), licensed MIT. It adds 63 tokens to every session and 1,502 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
test-fix
Diagnose and minimally fix a failing test, then re-run.
code-coherence-review
Reviews code and plans for logic coherence, correctness, and structural integrity.
hatch3r-incident-response
Handles production incidents with structured triage, mitigation, and post-mortem. Use when responding to production issues, outages, or security incidents.
hatch3r-bug-fix
Step-by-step bug fix workflow. Diagnose root cause, implement minimal fix, write regression test. Use when fixing bugs, working on bug report issues, or when the user mentions a bug.
hatch3r-perf-audit
Profiles and optimizes application performance against defined budgets. Use when investigating performance issues, auditing performance budgets, or optimizing hot paths.
icode
A Chinese-language, end-to-end coding workflow with separate steps for planning, review, approval, implementation, checking, and final audit. It also includes optional requirement drafting and log-based cause analysis.