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 rules/lazyagi/lazymind/coding-standardsgit clone --depth 1 https://github.com/LazyAGI/LazyMindWhat 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 | $0.01620 | $0.01620 |
| Opus 5 | $0.00810 | $0.00810 |
| Sonnet 5 | $0.00324 | $0.00324 |
| Haiku 4.5 | $0.00162 | $0.00162 |
Grade A, and why
coding-standards scanned grade A 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
### 推荐：容器内 curl 模拟 How it starts
The opening of the file, as written. The whole thing — 126 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Coding Standards
Comments
- Write all comments in English.
- Use clear, concise language.
Python
- Prefer single quotes (
') for strings when both are valid. - Example:
msg = 'hello'notmsg = "hello". - exec
make lintafter coding, runsource /Users/macro/lazyllm/bin/activateif environment is not ready - add
${PROJECT_ROOT_DIR}/algorigtm/lazyllmtoPYTHONPATHenv variable
Go
- use gofmt after modifying
Submodule
- Modidy
algorithm/lazyllmdirectly when you need to modify lazyllm. DO NOT modifysite-pickages
Debug
When debugging, do not infer root causes from code alone. First collect evidence by adding logs, reproducing the issue, simulating requests, inspecting database records, reviewing application logs, and tracing the actual execution path. Every diagnosis should be supported by runtime evidence. Avoid speculative fixes that are not verified against observed behavior.
Refactoring Role
Whenever you refactor code, you must clean up the old implementation immediately. Remove dead code, unused functions, unused variables, obsolete comments, redundant logic, and superseded implementations. Do not keep both old and new versions of the same logic unless backward compatibility is explicitly required. The codebase should be cleaner after every refactor, not accumulate technical debt.
Tool Descriptions vs System Prompt Guidance
All behavioral instructions for a tool (when to use it, how to call it, what types it accepts, what happens after it runs) belong in the tool function's docstring. Do NOT add corresponding guidance strings to algorithm/lazymind/chat/engine/prompts/guidance.py or inject them into the system prompt.
The LLM reads tool schemas (function name + docstring) directly at inference time. Guidance strings in the system prompt duplicate this information, go stale, and add noise. Keep the docstring the single source of truth.
Guidance strings in guidance.py are reserved for cross-cutting concerns that are not specific to any single tool (e.g. image URL formatting, citation style, environment context).
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 · 126 lines · 1,620 tokens per session scan A db82761f585f
coding-standards is a cursor rule published in the GitHub repository LazyAGI/LazyMind (74 stars, last pushed 2d ago), licensed Apache-2.0. It adds 1,620 tokens to every session, about $0.0081 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other cursor rules, from other repositories
skill-roots
Skill root ownership and which copy Cursor Agent must follow.
agent-instructions
Agent behavior guidelines.
token-efficiency
Token efficiency rules - eliminate waste in AI output, enforce read-before-write, prevent iteration cycles.
checkpoint
Write a committed handoff checkpoint so a fresh session can resume with zero loss. Use when crossing the 40% context line, on every stage transition, and at session end. Writes .trace/checkpoints/ - .md while the agent is still sharp.
repository-layout
Canonical map of repo folders and docs.
conservative-file-creation
Check for existing files before creating. Be conservative about new files and code.