Explore area
Skill huzaifa525/claude-code-optimizer/templates/.claude/skills/explore-area
Use when you need to understand a module, feature, or directory before making changes.From its SKILL.md
npx -y skills add huzaifa525/claude-code-optimizer --skill explore-areaAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- 9 stars9 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
- runs commandsInstructs the agent to run 3 commands, including `find [target] -type f -not -path '*/node_modules/*' -not -path '*/.git/*' | head -50` and 2 more.
SKILL.md
2.8 KB, 636 tokens by cl100k_base, as published. Nobody here has run it
Explore $ARGUMENTS using progressive disclosure to minimize token cost.
Progressive Disclosure (3 Layers)
Layer 1: File Index (cheapest — start here)
Get the file list WITHOUT reading contents:
# List all files in the target area
find [target] -type f -not -path '*/node_modules/*' -not -path '*/.git/*' | head -50
Or use Glob to find relevant files. From this list alone, identify:
- Entry points (index., main., app.*)
- Test files (.test., .spec.)
- Config files
- File count and general structure
Decision point: Do you now have enough context? If yes, skip to Return Format. If no, proceed to Layer 2.
Layer 2: Entry Points & Signatures (moderate cost)
Read ONLY entry points and key files. Use targeted reads:
- Read index/main files to understand exports and module structure
- Use Grep to find function signatures, class definitions, type exports
- Map imports/exports to understand dependency graph
# Find exports
grep -rn "export" [target]/index.* 2>/dev/null
# Find class/function definitions
grep -rn "^export \(class\|function\|const\|interface\)" [target]/ 2>/dev/null
Decision point: Do you understand the module's API and structure? If yes, skip to Return Format. If no, proceed to Layer 3.
Layer 3: Deep Dive (expensive — only when needed)
Read full file contents, but only for files that are:
- Directly relevant to the user's upcoming changes
- Complex enough that signatures alone don't explain behavior
- Containing patterns the user needs to follow
Use offset and limit for large files — read the relevant section, not the whole file.
Return Format
## Area: [name]
### Structure ([X] files)
- [file] → [purpose, 5 words max]
### Key Files (read these before editing)
- [file] → [what it does and why it matters]
### Dependencies
- Imports from: [external modules used]
- Exported to: [who consumes this module]
### Patterns to Follow
- [naming, structure, style patterns observed]
- [reference file] → follow this as template for new code
### Tests
- [test approach, test file locations, how to run]
### Gotchas
- [anything unusual, legacy code, known quirks]
- [things that look wrong but are intentional]
Rules
- Start at Layer 1. Only go deeper when needed.
- Never read ALL files in a directory — that defeats the purpose.
- Focus on what someone needs to know BEFORE making changes.
- Be concise. The user will read the files themselves when editing.
What ships with it
Read from the repository
Just SKILL.md. No reference files, no scripts.