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/serialx/vibecore/agents-mdgit clone --depth 1 https://github.com/serialx/vibecoreWhat 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 | $0.00662 | $0.00662 |
| Opus 5 | $0.00331 | $0.00331 |
| Sonnet 5 | $0.00132 | $0.00132 |
| Haiku 4.5 | $0.00066 | $0.00066 |
Grade A, and why
vibecore AGENTS.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 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.
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 — 43 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Repository Guidelines
Project Structure & Module Organization
src/vibecoreholds the app (main.py), Flow Mode orchestration (flow.py), and subpackages for agents, tools, widgets, session, models, and mcp.- Entry points are
cli.pyandmain.py; defaults and prompts live insettings.pyandprompts/. - Flow and multi-agent samples are under
examples/; reference docs stay indocs/. - Tests mirror runtime modules in
tests/, with fixtures intests/fixturesand harness helpers intests/_harness.
Build, Test, and Development Commands
uv sync— provision the dev environment againstuv.lock.uv run vibecore— launch the TUI locally; (NEVER RUN THIS)uv run vibecore --print -p 'prompt'— launch vibecore in pipe mode. non-TUI. (USE THIS INSTEAD OF TUI)uv run pytest …— run the suite or scoped folders liketests/ui.uv run ruff check ./uv run ruff format .— lint and auto-format.uv run pyright— type-check before requesting review.
Coding Style & Naming Conventions
- Python 3.11, 4-space indentation, 120-character lines, double quotes via Ruff.
- Use snake_case for modules, functions, variables; PascalCase for classes and TypedDicts.
- Keep public APIs typed—
py.typedmeans downstream users rely on annotations. - Place new tools, widgets, or agents beside existing peers to keep modules focused.
Testing Guidelines
- pytest runs in strict asyncio mode; decorate async tests with
pytest.mark.asyncio. - Name files
test_*.py, reuse fixtures, and mirror new runtime code with matching tests. - UI changes touching message widgets must refresh
tests/ui/test_message_snapshots.py. - Run full
uv run pytestplus targeted suites before opening a PR.
Textual & UI Development Tips
- Consult Textual source in
.venv/lib/python3.11/site-packages/textual/when extending widgets—docs lag the implementation. - Textual ignores CSS
@keyframes; use widgetanimate()helpers for motion. - Streaming handlers in
handlers/expect incremental updates—ensure new widgets cooperate with that flow.
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 · 43 lines · 662 tokens per session scan A 518ebd6568e4
vibecore AGENTS.md is an instructions file published in the GitHub repository serialx/vibecore (23 stars, last pushed 8mo ago), licensed MIT. It adds 662 tokens to every session, about $0.0033 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 2026-08-30.
Other instructions, from other repositories
mini-swe-agent AGENTS.md
AGENTS.md instructions for SWE-agent/mini-swe-agent, covering mini-swe-agent overview, style guide, bad, good and test style.
mini-swe-agent copilot-instructions.md
Copilot instructions for SWE-agent/mini-swe-agent, covering style guide, test style, bad and good.
tact AGENTS.md
Instructions for clabby/tact, covering working agreements, common dev commands, code, secrets and documentation.
aegis AGENTS.md
Instructions for apiad/aegis, covering agents, running, know-how, package management and layout.
q-code AGENTS.md
Instructions for v833/q-code, covering q-code 项目协作说明, 项目概览, 环境与工具, 常用命令 and cli 子命令.
mini-swe-agent CLAUDE.md
Claude Code instructions for SWE-agent/mini-swe-agent, a project described as: The 100 line AI agent that solves GitHub issues or helps you in your command line. Radically simple, no huge configs, no giant monorepo—but scores >74% on SWE-bench verified!