idiomatic-by-default
Take the community-standard path for the language, framework, or ecosystem unless there is an articulable reason not to, and don't escalate a decision the…
Skill by kasofsk
Chalk Labs
Need help deploying idiomatic-by-default?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/kasofsk-idiomatic-by-default
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: backend
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/idiomatic-by-default
What's inside
- Skill name
idiomatic-by-default- Description
- Take the community-standard path for the language, framework, or ecosystem unless there is an articulable reason not to, and don't escalate a decision the idiom already answers. Use when choosing a library, project layout, API shape, or tooling approach, when about to hand-roll something the ecosystem already provides, or before asking the operator to decide a question the ecosystem has settled.
- License
- MIT
- Instructions
- 44 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# Idiomatic by Default ## Core principle Virtually always do things the idiomatic way. We never want to get caught with our pants down having built something more complicated or more brittle when an idiomatic path was available to us. ## How we get there ### 1. The idiomatic path is the default path For any given language, framework, or ecosystem, the community-standard way of doing something is the presumptive choice. Idioms are battle-tested: they encode years of collective debugging, and every future reader (human or agent) already knows how they work. ### 2. Deviation requires a very good reason This is not an absolute rule — we are allowed to leave the idiomatic path. But if we do, we should have *very good reasons*, and those reasons should be articulable. "It seemed cleaner to me" or "I didn't check what the standard approach was" don't qualify. The failure mode we're guarding against is discovering, after the fact, that we hand-rolled something complicated and brittle where a well-worn idiom existed all along. ### 3. Don't escalate decisions an idiom already answers When making architectural or technical decisions — including deciding whether to send a question back to the human operator — check whether one of the options is the tried-and-true idiomatic path. If it is, **default to it** rather than forcing the operator to make that call. Operator attention is a scarce resource; spend it on genuinely open questions, not on choices the ecosystem has already settled. ## Why this matters Idioms are a form of pre-paid knowledge: choosing them means inheriting the ecosystem's testing, tooling, documentation, and reader familiarity for free. Non-idiomatic code taxes every future change — and every escalation of an already-settled question taxes the operator. Both are costs we should only pay deliberately, never by accident.
From kasofsk/blessed-practices, MIT.
More from kasofsk/blessed-practices
- 0.524score
comments-describe-the-code
kasofsk
Write comments about the code itself, never about the prompt, conversation, or moment that produced it — and treat a long justifying comment as a signal the…
SkillBFreshness: BClaude Code +9 - 0.524score
design-with-known-limits
kasofsk
Design systems with explicit correctness limits, failure modes, and operational assumptions instead of automatically engineering every weakness away.
SkillBFreshness: BClaude Code +9 - 0.524score
diminishing-returns
kasofsk
Pursue the highest-value work first and stop when the expected value of the next improvement no longer justifies its full cost.
SkillBFreshness: BClaude Code +9 - 0.524score
domain-modelling
kasofsk
Shape the contents of the domain core — ubiquitous language, aggregates as consistency boundaries, entities versus value objects, making illegal states…
SkillBFreshness: BClaude Code +9 - 0.524score
fix-the-assumption-not-the-hack
kasofsk
When a new feature invalidates an assumption the current design rests on, rethink the pattern rather than land a tweak that technically satisfies the request.
SkillBFreshness: BClaude Code +9 - 0.524score
illegal-states-unrepresentable
kasofsk
Make illegal states unrepresentable with precise types, discriminated variants, and boundary parsing.
SkillBFreshness: BClaude Code +9 - 0.524score
layering
kasofsk
Decide where code goes and which way its dependencies point — inward-pointing source dependencies, what the domain, application, adapter and infrastructure…
SkillBFreshness: BClaude Code +9 - 0.524score
modular-and-layered-code
kasofsk
Architect so a future agent can act at a single layer — strict tested contracts between layers, testing concentrated at domain boundaries, pure logic kept…
SkillBFreshness: BClaude Code +9 - 0.524score
push-back-and-verify-assumptions
kasofsk
Treat the operator as a fellow engineer — challenge naive or wrong premises, state assumptions explicitly, keep asking until they are right, and stop rather…
SkillBFreshness: BClaude Code +9 - 0.524score
write-the-present-not-the-diff
kasofsk
Write prose that states what is true now, without narrating the correction, the superseded claim, or the path that got there.
SkillBFreshness: BClaude Code +9 - 0.494score
dependencies
kasofsk
Add, remove, upgrade, or pin a dependency with the ecosystem's package manager instead of hand-editing the manifest.
SkillBFreshness: BClaude Code +9
Signals
- Stars
- 0
- Forks
- 1
- Open issues
- 0
- Last commit (this folder)
- Aug 20, 2026 (2 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/kasofsk/blessed-practices
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from kasofsk
- 0.524score
comments-describe-the-code
kasofsk
Write comments about the code itself, never about the prompt, conversation, or moment that produced it — and treat a long justifying comment as a signal the…
SkillBFreshness: BClaude Code +9 - 0.524score
design-with-known-limits
kasofsk
Design systems with explicit correctness limits, failure modes, and operational assumptions instead of automatically engineering every weakness away.
SkillBFreshness: BClaude Code +9 - 0.524score
diminishing-returns
kasofsk
Pursue the highest-value work first and stop when the expected value of the next improvement no longer justifies its full cost.
SkillBFreshness: BClaude Code +9 - 0.524score
domain-modelling
kasofsk
Shape the contents of the domain core — ubiquitous language, aggregates as consistency boundaries, entities versus value objects, making illegal states…
SkillBFreshness: BClaude Code +9 - 0.524score
fix-the-assumption-not-the-hack
kasofsk
When a new feature invalidates an assumption the current design rests on, rethink the pattern rather than land a tweak that technically satisfies the request.
SkillBFreshness: BClaude Code +9 - 0.524score
illegal-states-unrepresentable
kasofsk
Make illegal states unrepresentable with precise types, discriminated variants, and boundary parsing.
SkillBFreshness: BClaude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get idiomatic-by-default running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
idiomatic-by-default skill. Agent Mart. https://agentmart.dev/skills/kasofsk-idiomatic-by-default (last updated 2026-10-24)