expectations
A framework for capturing project decisions, gotchas, and patterns in documentation.
Install
mkdir -p .claude/skills/expectations && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/3148" && unzip -o skill.zip -d .claude/skills/expectations && rm skill.zipInstalls to .claude/skills/expectations
Activation
This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.
Capture learnings, gotchas, and architectural decisions into the right project documentation while context is fresh. Use when capturing learnings, documenting gotchas, recording architectural decisions, or deciding where a piece of knowledge should live. Triggers on "document this", "remember this pattern", "what should I know about", or after completing significant features.Key capabilities
- →Identify project-specific gotchas
- →Record architectural trade-offs
- →Document effective coding patterns
- →Capture tool configuration insights
How it works
Developers identify learnings at the end of a feature, then categorize them by type to determine if they belong in CLAUDE.md, ADRs, or project documentation.
Inputs & outputs
When to use expectations
- →Document a project-specific gotcha
- →Record architectural rationale for a design decision
- →Capture an effective coding pattern
- →Document domain-specific knowledge
About this skill
Expectations: Capturing Learnings
Route each discovery to the artifact that owns its truth. Do not turn one project's architecture, delivery process, or tooling conventions into universal guidance.
Reusable Principle or Local Policy?
- Reusable principle — applies across projects without relying on one repository's names, paths, tools, or release process. Improve the relevant global skill.
- Local policy — depends on this product's domain, architecture, risk posture, workflow, or infrastructure. Record it in repository agent guidance or canonical project documentation.
When in doubt, make the narrower local update. Reuse should be earned by evidence from more than one project, not inferred from one successful implementation.
Documentation Framework
At the end of every significant change, ask: "What do I wish I'd known at the start?"
Capture a learning if ANY of these are true:
- Would save future developers significant time
- Prevents a class of bugs or errors
- Reveals non-obvious behavior or constraints
- Captures architectural rationale or trade-offs
- Documents domain-specific knowledge
- Identifies effective patterns or anti-patterns
- Clarifies tool setup or configuration gotchas
Do not duplicate what the repository already records: current code structure, Git history, or behavior obvious from source and tests.
Types of Learnings to Capture
- Gotchas: Unexpected behavior discovered (e.g., "API returns null instead of empty array")
- Patterns: Approaches that worked particularly well
- Anti-patterns: Approaches that seemed good but caused problems
- Decisions: Architectural choices with rationale and trade-offs
- Edge cases: Non-obvious scenarios that required special handling
- Tool knowledge: Setup, configuration, or usage insights
Where Each Learning Goes
| Learning | Destination | Why |
|---|---|---|
| Cross-project practice with evidence beyond one repository | Relevant global skill | Reusable guidance stays provider- and project-neutral |
| Project-specific architecture, delivery rule, tool setup, or gotcha | Repository agent guidance or canonical project documentation | Local policy stays with the project that owns it |
| Domain term, definition, alias, or naming rule | Owning bounded context's authoritative glossary | Vocabulary has one context-scoped authority |
| Accepted architecture, dependency, platform, or build-versus-adopt decision | Repository's accepted ADR location | Durable decisions retain rationale and rejected alternatives |
| Observable behavioral constraint or regression | Source code and executable tests | The maintained behavior proves the rule |
| In-flight blocker, sequencing choice, or scope change | Current status artifact or active plan | Temporary delivery knowledge expires with the work |
| User-facing behavior, setup, API usage, or operations | Maintained product, package, or operational documentation | Readers find current truth at the owning surface |
If a temporary artifact reveals a lasting constraint, promote the constraint to its owner before deleting the artifact. Git history remains the archive.
Documentation Format
#### Gotcha: [Descriptive Title]
**Context**: When this occurs
**Issue**: What goes wrong
**Solution**: How to handle it
// CORRECT - Solution
const example = "correct approach";
// WRONG - What causes the problem
const wrong = "incorrect approach";
Keep entries scannable: a future reader should grasp context, issue, and solution in under ten seconds.
Communication
- Be explicit about trade-offs in different approaches
- Explain the reasoning behind significant design decisions
- Flag any deviations from guidelines with justification
- Suggest improvements that align with these principles
- When unsure, ask for clarification rather than assuming
- State explicitly whether a recommendation is a reusable principle or local policy
When not to use it
- →Documenting code structure
- →Recording git history
- →Capturing information derivable from code
Prerequisites
Limitations
- →Requires manual identification of documentation-worthy insights
- →Must not duplicate existing repository records
How it compares
This method forces documentation of non-obvious constraints and rationale rather than just describing code functionality.
Compared to similar skills
expectations side by side with the closest alternatives in the catalog.
| Skill | Installs | Updated | Safety | Difficulty |
|---|---|---|---|---|
| expectations (this skill) | 1 | 2mo | No flags | Beginner |
| deepwiki-rs | 25 | 9mo | Review | Intermediate |
| codex-cli-bridge | 9 | 9mo | Review | Intermediate |
| skill-development | 17 | 9mo | Review | Intermediate |
Try saying
Example prompts that trigger this skill in your AI assistant.
More by citypaul
View all by citypaul →You might also like
deepwiki-rs
sopaco
AI-powered Rust documentation generation engine for comprehensive codebase analysis, C4 architecture diagrams, and automated technical documentation. Use when Claude needs to analyze source code, understand software architecture, generate technical specs, or create professional documentation from any programming language.
codex-cli-bridge
alirezarezvani
Bridge between Claude Code and OpenAI Codex CLI - generates AGENTS.md from CLAUDE.md, provides Codex CLI execution helpers, and enables seamless interoperability between both tools
skill-development
anthropics
This skill should be used when the user wants to "create a skill", "add a skill to plugin", "write a new skill", "improve skill description", "organize skill content", or needs guidance on skill structure, progressive disclosure, or skill development best practices for Claude Code plugins.
skill-writer
pytorch
Guide users through creating Agent Skills for Claude Code. Use when the user wants to create, write, author, or design a new Skill, or needs help with SKILL.md files, frontmatter, or skill structure.
openapi-spec-generation
wshobson
Generate and maintain OpenAPI 3.1 specifications from code, design-first specs, and validation patterns. Use when creating API documentation, generating SDKs, or ensuring API contract compliance.
korean-skill-creator
clwmfksek
한글 기반 클로드 스킬 자동 생성 도구. 사용자가 "클로드 스킬을 만들어줘" 또는 "[요구사항] 스킬 만들어줘"라고 요청할 때 사용. Progressive disclosure 원칙을 따르는 한글 문서 구조(SKILL.md + references/)를 자동으로 생성하고, 실전 예시를 포함한 일관성 있는 스킬 템플릿을 제공.