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 instructions/vironnimo/vbot/claude-mdgit clone --depth 1 https://github.com/Vironnimo/vbotWhat 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 | $0.03305 | $0.03305 |
| Opus 5 | $0.01653 | $0.01653 |
| Sonnet 5 | $0.00661 | $0.00661 |
| Haiku 4.5 | $0.00331 | $0.00331 |
Grade A, and why
vbot CLAUDE.md 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 yesterday.
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 — 97 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
You work in this repo with your full capabilities and your usual workflow — directly on main, using every tool, skill, and subagent you'd normally reach for. Nothing here narrows your agency; work to your full potential.
"Not part of the orchestrator" means only this: you are not a managed node in the vorch orchestrator system — no orchestrator assigns or reviews your work, and its roles (builder, tester, reviewer, …) are not yours to call; you do that work yourself. It is not a limit on the subagents you spawn for your own tasks. What you share with that system is its resources — .vorch/PROJECT.md, .vorch/GLOSSARY.md, the domain maps, .vorch/DESIGN.md, .vorch/FLAGGED.md, the workflows, the skills, and the project's conventions — which you use and keep current in return. That's the whole relationship: consume the resources, maintain them, follow the rules.
Talking to the User
The user reads no code. All user-facing communication — discovery, plan review, decisions, escalations, summaries — is in product language: behavior, capabilities, consequences, domains. No file paths, function names, or code identifiers unless they are part of the product interface (a CLI command, an API endpoint, a config option).
Decisions — surface by whether it's a real choice, not by whether the user can feel it. A change built to be imperceptible still has sub-decisions worth seeing; "the user won't feel it" is never a reason to bury one.
- Ask first when the user has a stake in the outcome — product behavior, or a trade-off they'd want to weigh in on: present it in the options format below, then wait.
- Decide, then disclose in one line when it's a real choice among defensible options the user would likely leave to good practice (a 0.5s vs 0.8s debounce, a threshold, a library): make the call and show it — what you chose, the alternative, and why, in one overridable line.
- Settle silently only when no alternative exists that the user could hold a view on — naming, code structure below domain level, where code lives.
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.
- yesterday First seen · 97 lines · 3,305 tokens per session scan A dec4954ff876
vbot CLAUDE.md is an instructions file published in the GitHub repository Vironnimo/vbot (1 stars, last pushed 2d ago), licensed Apache-2.0. It adds 3,305 tokens to every session, about $0.0165 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 instructions, from other repositories
dograh AGENTS.md
AGENTS.md instructions for dograh-hq/dograh, covering dograh - project overview, project structure, tech stack, local development and environment configuration.
MemMachine AGENTS.md
AGENTS.md instructions for MemMachine/MemMachine, covering memmachine agent guide, repository orientation, environment setup, build, lint, and test commands and python (repo root).
bub AGENTS.md
AGENTS.md instructions for bubbuild/bub, covering repository guidelines, project structure & module organization, build, test, and development commands, coding style & naming conventions and testing guidelines.
dify AGENTS.md
AGENTS.md instructions for langgenius/dify, covering agents.md, repository gotchas and frontend workflow.
dify CLAUDE.md
Claude Code instructions for langgenius/dify, a project described as: Build Agentic workflows, RAG pipelines, with rich AI model and tool support on one collaborative workspace. Deploy on cloud, VPC, or self-hosted, so teams move from prototype to production without rebuilding the stack.
minion CLAUDE.md
Instructions for femto/minion, covering 🧠 记忆存储, 系统架构记忆 and 开发流程记忆.