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/sy159/snowgo/agents-mdgit clone --depth 1 https://github.com/sy159/snowgoWrote 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/instructions/sy159/snowgo/agents-md)<a href="https://agentmods.dev/instructions/sy159/snowgo/agents-md"><img src="https://agentmods.dev/badge/instructions/sy159/snowgo/agents-md.svg" alt="Measured on agentmods" 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 | $0.00794 | $0.00794 |
| Opus 5 | $0.00397 | $0.00397 |
| Sonnet 5 | $0.00159 | $0.00159 |
| Haiku 4.5 | $0.00079 | $0.00079 |
Grade A, and why
snowgo 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 — 37 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Go Version: 1.26+ (toolchain: go1.26.7) Status: Active — single source of truth for engineering decisions.
Document Index
| Document | Scope |
|---|---|
| CODING.md | Naming, error handling, logging, context, validation, concurrency |
| ARCHITECTURE.md | Database design, transactions, caching, query optimization |
| OPERATIONS.md | Security, testing, review checklist |
Core Rules
- Never manually edit
internal/dal/model/orinternal/dal/query/. - Service owns transaction boundaries. DAO methods accept caller-provided
*query.Queryand never open transactions themselves. Multi-table mutations and audited business mutations that require operation logs useWriteQuery().Transaction(). Independent single-table writes may use a non-transactional*query.Querywhen atomic cross-table consistency is not required. Operation logs for audited business mutations are written synchronously within the same transaction through transaction-aware contracts or DAO methods. - All DAO methods that may participate in a transaction accept
*query.Queryas a parameter. The DAO does not care whether it runs in a transaction — the Service layer decides: passtxinsideTransaction(), passrepo.Query()orrepo.WriteQuery()outside. There is only one method per DAO operation — no separateTransaction*Xxxvariants. - Cache invalidation happens after DB commit, never inside a transaction.
- Admin endpoints require
JWTAuth()after login. AddPermissionAuth(constant.PermXxx)for privileged management or scoped business data endpoints. Login-only endpoints (current user permissions, server info, allowed dictionary lookups) must be documented in route comments. Middleware ininternal/router/middleware/auth.go. - Service dependencies follow package boundaries: same-package services inject concrete
*XxxService; cross-package service calls use stable contracts frominternal/service/admin/contractwhen the dependency is shared infrastructure capability, and consumer-local interfaces only for narrow one-off domain dependencies. Do not call business Services inside an open transaction; only transaction-aware infrastructure contracts that receive the caller's*query.Queryare allowed. - Use
xlogger.InfofCtx/xlogger.ErrorfCtxfor all logging. - Define sentinel errors in Service using
pkg/xerror/codes; compare witherrors.Is. - API layer validates all input before reaching Service.
- No
panic()in API/Service/DAO for business errors. created_atmandatory for all tables; soft delete (is_deleted tinyint(1) DEFAULT 0+deleted_at DATETIME(6) DEFAULT NULL) is optional, decide per business need. Use for tables requiring audit/compliance/user undo (e.g., orders). Skip for high-volume logs and junction tables.- Run tests according to change scope before declaring complete: small, localized changes run affected package tests; broad/shared behavior changes run
go test ./.... Always runmake lint. Use coverage commands only when coverage is the explicit goal. Integration tests use-tags=integration;make test-integrationcovers pkg and Service integration tests and requires Redis/RabbitMQ plusMYSQL_DSNfor Service tests. - Core and complex code must have comments. Update README / AGENTS docs alongside code changes.
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 · 37 lines · 794 tokens per session scan A a0966f9b8e93
snowgo AGENTS.md is an instructions file published in the GitHub repository sy159/snowgo (23 stars, last pushed 7d ago), licensed MIT. It adds 794 tokens to every session, about $0.0040 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
vc-gin-api AGENTS.md
Instructions for bg-vc/vc-gin-api, covering vc-gin-api agent guide, product rule: simple and change workflow.
jwx AGENTS.md
AGENTS.md instructions for lestrrat-go/jwx, covering agents.md, for module consumers, go version, goexperiment and module path vs physical layout.
picoaide AGENTS.md
Instructions for picoaide/picoaide, covering claude.md, 项目概述, 构建与运行, 构建命令 and ⚠️ 重要:必须全量编译.
jwx CLAUDE.md
Claude Code instructions for lestrrat-go/jwx, a project described as: Complete implementation of JWx (Javascript Object Signing and Encryption/JOSE) technologies for Go. #golang #jwt #jws #jwk #jwe.
awesome-go AGENTS.md
AGENTS.md instructions for avelino/awesome-go, covering awesome-go · llm contribution guide, project snapshot, when modifying the awesome list, coding guidelines and testing & validation.
ollama CLAUDE.md
Claude Code instructions for ollama/ollama: See AGENTS.md for the shared agent instructions for this repository.