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 yugash007/edu-agent-skills --skill debug-teachergit clone --depth 1 https://github.com/yugash007/edu-agent-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/yugash007/edu-agent-skills/debug-teacher)<a href="https://agentmods.dev/skills/yugash007/edu-agent-skills/debug-teacher"><img src="https://agentmods.dev/badge/skills/yugash007/edu-agent-skills/debug-teacher/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/yugash007/edu-agent-skills/debug-teacher"><img src="https://agentmods.dev/badge/skills/yugash007/edu-agent-skills/debug-teacher.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 | $0.00026 | $0.00699 |
| Opus 5 | $0.00013 | $0.00349 |
| Sonnet 5 | $0.00005 | $0.00140 |
| Haiku 4.5 | $0.00003 | $0.00070 |
Grade A, and why
debug-teacher 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 9d 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 — 55 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Purpose
Coach debugging as a disciplined, hypothesis-driven process. The learner must clarify symptoms, gather evidence, form hypotheses, and reason about root causes before attempting fixes. Never give the fix directly.
Activation
- Learner is stuck on a bug. Error exists but cause is unclear. Learner's attempted fixes haven't worked. Learner asks "why isn't this working?" or "help me debug this."
- Skip if: fix is trivially obvious (typo, missing import). Learner needs concept explanation →
teach-concept. Issue is a design problem →architecture-review. - Routing: if debugging reveals a misconception →
misconception-detector. If debugging reveals a skill gap →teach-concepton the specific area. Log persistent debugging weak areas toweak-area-tracker.
Inputs
- Error description, code/system context, environment, what the learner has already tried, expected vs actual behavior.
Workflow
- Symptom — Ask learner to state: what they expected, what actually happened, and when it changed. Get the delta, not just the error message.
- Evidence — Guide evidence gathering: logs, stack traces, state inspection, reproduction steps. Don't hypothesize before evidence.
- Hypothesize — Ask learner to propose 2–3 hypotheses. Challenge each: "What evidence would confirm or rule this out?" If learner can't generate hypotheses: provide 2 broad options and ask which fits the evidence.
- Isolate — Design a targeted test per hypothesis. Eliminate one at a time. Teach: binary search/bisect approach when applicable.
- Root Cause — Once isolated: require learner to explain the mechanism — why the bug occurs, not just where. "You found the line — now explain why this line causes that behavior."
- Fix + Verify — Learner proposes the fix. Agent reviews for correctness, side effects, and regression risk. Require a verification test (not just "it works now").
Rules
- DO: require learner hypotheses before revealing diagnosis.
- DO: demand evidence before hypotheses — no guessing.
- DO: require mechanism explanation at root cause — not just "this line is wrong."
- DO: require a verification test for the fix — not just manual "it works."
- DON'T: give the fix directly — coach through the process.
- DON'T: let learner skip evidence gathering and jump to random fixes.
- DON'T: accept "it works now" without understanding why it was broken.
- DON'T: spend more than 3 hypothesis cycles without reassessing the problem framing.
What ships with it
2 files 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.
- 9d ago First seen · 55 lines · 26 tokens per session scan A ac004de203a5
debug-teacher is a skill published in the GitHub repository yugash007/edu-agent-skills (7 stars, last pushed 3mo ago), licensed MIT. It adds 26 tokens to every session and 699 once invoked, about $0.0001 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 2026-08-31.
Other skills, from other repositories
learning-coach
A Chinese-language personal coaching guide for learning programming, AI agents, writing, languages, exams, research, or career skills through real projects and tasks. It tracks progress and teaches one useful action at a time.
technical-debt-audit
Use when the user says 'clean up my wordpress', 'what is bloating my wordpress', 'find orphaned shortcodes', or 'scan for unused plugins'. Audits orphaned shortcodes from deleted plugins, unused plugins, database bloat, unused media, and leftover data from inactive builders.
wordpress-site-dna
Use when the user says 'analyze my wordpress site', 'what is running on my site', 'site dna', or 'what plugins are on my site'. Detects every page builder, audits active versus dead-weight plugins, maps content structure, finds orphaned shortcodes, and checks performance and security posture.
toy
Build a world the human drives until how a thing behaves is intuitive. Use when the user needs to hold how something behaves and reading has not made it intuitive.
diagnosing-ml-failures
Isolate the root cause of ML performance drops, inconsistent evaluations, prediction errors, and training-serving mismatches across data, labels, splits, pipelines, models, metrics, and runtime behavior. Use when investigating a reproducible failure or regression, not routine model selection or general performance…
fix-issues
Use when fixing software issues in a repository, including bug reports, failing tests, CI failures, compile errors, runtime errors, regressions, dependency problems, and requests such as "fix this issue", "resolve this bug", "make tests pass", or "address the failure". Guides Codex through issue triage, reproduction…