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/liyi-run/liyi/agents-mdgit clone --depth 1 https://github.com/liyi-run/liyiWhat 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 | $0.04890 | $0.04890 |
| Opus 5 | $0.02445 | $0.02445 |
| Sonnet 5 | $0.00978 | $0.00978 |
| Haiku 4.5 | $0.00489 | $0.00489 |
Grade A, and why
liyi 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 3d 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 — 392 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
This repository is 立意 (Lìyì) — a convention and tooling project for making intent explicit, persistent, and reviewable in AI-assisted software development.
立意 is not an agent skill collection — it defines the practice and ships the tools (linter, spec format, agent skill template) that other projects adopt. One of its deliverables is an agent skill template that downstream repositories can use to bootstrap their own AGENTS.md and contributing workflows.
Before making any changes, read the contributing guide that matches the predominant language of the current user session:
- 中文会话 → docs/contributing-guide.zh.md
- English session → docs/contributing-guide.en.md
If the session language is unclear, default to the Chinese version.
The contributing guide covers project structure, content workflows, code style, commit message conventions, AIGC policy compliance, and the validation checklist. All of those rules are authoritative and must be followed.
Code style (quick reference)
For Rust code:
- Conform to official rustfmt style.
- Keep free from Clippy lints.
Before committing, run make lint (rustfmt check, Clippy, and the linter's
own self-check) and ensure it passes, or make verify to also run the test
suite. Both gates abort at the first failure and end with a success banner
(lint: all gates passed / verify: all gates passed), so you can confirm
success by the final line — no need to capture the exit code or redirect
output to a log.
If you are an AI agent, prefer the quiet variants make lint-short and
make verify-short. They run the exact same gates but suppress normal
output to conserve your context window: on success they print only a banner
(lint-short: all gates passed / verify-short: all gates passed), and on
failure they print a short tail of the failing gate's output plus the full
make <gate> target to re-run for complete diagnostics. Run that full target
when you need to see everything. Do not hand-roll make lint >log 2>&1; echo $? pipelines — the short variants already do this correctly.
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.
- 3d ago First seen · 392 lines · 4,890 tokens per session scan A 15a96fc32406
liyi AGENTS.md is an instructions file published in the GitHub repository liyi-run/liyi (5 stars, last pushed 3mo ago), licensed Apache-2.0. It adds 4,890 tokens to every session, about $0.0244 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
open-ontologies CLAUDE.md
Instructions for fabio-rovai/open-ontologies, covering open ontologies, ontology engineering workflow, generate, validate and load and reason.
kin GEMINI.md
Instructions for firelock-ai/kin, covering kin: read the repository from the graph, pick the tool by what you know, trust the envelope, not the emptiness, if the tools have no graph and a working order.
kin AGENTS.md
Instructions for firelock-ai/kin, covering agents.md and this repo's role.
gangsta AGENTS.md
Instructions for kucherenko/gangsta, covering contributing to gangsta, what this is, what will not be accepted, skill changes and the laws apply to contributors too.
featherspec memory-bank.instructions.md
FeatherSpec Memory Bank writing rules (thin loader).
agent-toolkit AGENTS.md
Instructions for eai-org/agent-toolkit, covering agent toolkit guide and docs.