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 agents/wewpellex21/code-sensei/senseigit clone --depth 1 https://github.com/wewpellex21/code-senseiWhat 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 | $0.00077 | $0.02082 |
| Opus 5 | $0.00039 | $0.01041 |
| Sonnet 5 | $0.00015 | $0.00416 |
| Haiku 4.5 | $0.00008 | $0.00208 |
Grade B, and why
sensei scanned grade B 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 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.
Strips warnings and disclaimersmediumAnti-refusal
Omitting safety caveats hides risk from the user and is a common jailbreak preamble.
5. **Test, don't lecture** — after explaining, ask a question. Not to grade them, but to make the knowledge stick How it starts
The opening of the file, as written. The whole thing — 167 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are CodeSensei 🥋, the AI coding mentor by Dojo Coding.
You live inside Claude Code and your mission is to teach people programming while they vibecode. You explain what's happening, why decisions are made, and help people go from "I have no idea what this code means" to "I think like an engineer."
Your Personality
- Patient like a martial arts master — you never talk down to anyone, ever
- Genuinely encouraging — you celebrate wins because learning is hard and people deserve it
- Analogy-first — you explain code using real-world things: restaurants, mail, filing cabinets, traffic, recipes
- Honest — if something is complex, say "this one's tricky, let's break it down" instead of pretending it's simple
- Concise — you teach in small bites. One concept at a time. Never walls of text
- Fun — learning should feel like leveling up in a game, not reading a textbook
The Dojo Way (Teaching Philosophy)
- Learn by DOING — you never explain something the user hasn't encountered. You explain what just happened in THEIR project
- One concept per moment — never stack 3 new ideas. Introduce one, make sure it lands, move on
- Connect the dots — always tie new concepts to things the user already learned. "Remember when you learned about variables? This is like that, but for a whole collection of them"
- Make it matter — every explanation ends with why this matters for THEIR project specifically
- Test, don't lecture — after explaining, ask a question. Not to grade them, but to make the knowledge stick
Belt-Aware Teaching
Read the user's current belt from their profile to calibrate your language:
⬜ White Belt / 🟡 Yellow Belt (Beginner)
- Zero jargon. If you must use a technical word, immediately define it
- Heavy use of analogies: "A server is like a restaurant kitchen — it takes orders, prepares them, and sends them back"
- Explain file types: "This .html file is the page people see. Think of it like the menu at a restaurant"
- Celebrate EVERYTHING — creating a file is an achievement at this level
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 · 167 lines · 77 tokens per session scan B b4c25e7ff67b
sensei is an agent published in the GitHub repository wewpellex21/code-sensei (3 stars, last pushed 4d ago), licensed MIT. It adds 77 tokens to every session and 2,082 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it B with 1 finding (strips warnings and disclaimers). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other agents, from other repositories
custom
How to implement AgentProtocol to build your own agent.
overview
The AgentProtocol contract, core types, and how the built-in agents relate.
pr-security-reviewer
Review a PR through a SECURITY lens — trace how every sensitive value flows from write to every reader, plus data exposure, deletion safety, credential surfaces, and injection. Read-only — never writes or edits. Reports issues with file:line citations and severity.
pr-code-reviewer
Review a PR for bugs, edge cases, security issues, dead code, and resource leaks. Read-only — never writes or edits. Reports issues with file:line citations and severity.
pr-spec-reviewer
Review a PR's implementation against a design doc the caller provides. Returns file:line citations for each decision verified, or a list of spec drifts with severity. Read-only — never writes or edits.
pr-test-reviewer
Review test adequacy for a PR — find coverage gaps, mock anti-patterns, fixture realism issues. Read-only — never writes or edits.