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 skills add Kevin-Liu-01/Agent-Machines --skill style-guidegit clone --depth 1 https://github.com/Kevin-Liu-01/Agent-MachinesWrote 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/skills/kevin-liu-01/agent-machines/style-guide)<a href="https://agentmods.dev/skills/kevin-liu-01/agent-machines/style-guide"><img src="https://agentmods.dev/badge/skills/kevin-liu-01/agent-machines/style-guide/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/kevin-liu-01/agent-machines/style-guide"><img src="https://agentmods.dev/badge/skills/kevin-liu-01/agent-machines/style-guide.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 5.1 | $0.00026 | $0.01374 |
| Opus 5 | $0.00013 | $0.00687 |
| Sonnet 5 | $0.00005 | $0.00275 |
| Haiku 4.5 | $0.00003 | $0.00137 |
Grade A, and why
style-guide 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 9d 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 — 163 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Dedalus Style Guide: Quick Reference
Hard Limits
- Functions: 70 lines max
- Files: 500 lines max (excluding inline tests)
- Nesting: 3 levels max
- Arguments: 5 max (excluding self/cls)
- PRs: 200 changed LOC max
I/O Convention
# I/O function: io: Io as first parameter
async def fetch_user(io: Io, user_id: str) -> User | None: ...
# Pure function: no io parameter
def compute_discount(user: User, cart: Cart) -> Decimal: ...
Types
- No
Any. Period. If you reach forAny, you have not modeled the domain. - No
objectas a type annotation. It isAnyin disguise. - For heterogeneous JSON, use
JSONValue,JSONObject,JSONPrimitive, orJSONArrayfromcore.types.json. These are proper recursive RFC 8259 types. There is no excuse fordict[str, object]ordict[str, Any]. - No
**kwargs: Anywithout justification. - Use typed structures (dataclass, TypedDict, Pydantic) instead of dicts.
- Pydantic at boundaries, plain types internally.
- Replace anonymous tuples with frozen dataclasses or named types.
Functions
- Early returns instead of nesting
- One thing per function: if you need "and", split it
- Named parameters at call sites:
fetch_user(io=io, user_id=uid) - Return value in own variable for debuggability:
result = ...; return result - No chaining construction + method:
Foo(x).bar()hides the instance. Split intofoo = Foo(x)thenresult = foo.bar(). - Split compound assertions:
assert a; assert bnotassert a and b
Error Handling
- Specific exceptions with domain context:
ChargeError(org_id, reason) - Never catch bare
Exceptionwithout# noqa: BLE001 - Never swallow with
pass - Namespace pattern:
BillingError.ChargeError - Result types for domain logic, try/except at boundaries
No Fallbacks
# Bad: silent, undebuggable
model = request.model or config.default_model or "gpt-4"
# Good: explicit precedence with early returns
def get_model(request, config):
if request.model:
return request.model
if config.default_model:
return config.default_model
raise ValueError("model is required")
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.
- 9d ago First seen · 163 lines · 26 tokens per session scan A b8bbd6b5435c
style-guide is a skill published in the GitHub repository Kevin-Liu-01/Agent-Machines (29 stars, last pushed today), licensed MIT. It adds 26 tokens to every session and 1,374 once invoked, about $0.0001 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 2026-09-03.
Other skills, from other repositories
coding.make_function_private
Identify private / public functions in a file and rename them with an underscore or not.
code-review-python
Provides Python-specific code review rules for the dh code-reviewer agent. Activates on pyproject.toml or .py file detection — enforces uv, ruff, ty, pytest, type annotation, error handling, and Python 3.11+ idioms including pathlib, match statements, and modern union syntax.
coding_qa.review
Review Python files for bugs, suggest fixes, and provide test cases.
verify-language
Language-specific verification for Python, TypeScript/JavaScript, and Go. Checks type safety, language idioms, and best practices. Use when asked to "verify language", "check types", or for language-specific checks.
python-refactoring
Analyze Python code for refactoring opportunities, complexity, dead code, security issues, modernization, test coverage gaps, and package-level structure — before proposing or making refactoring changes to Python code.
knowledge-audit
Review and clean up stored memories — find duplicates, contradictions, stale entries, and consolidate.