hermeticormus/code-comprehension-skills · Archived
Understand an unfamiliar codebase fast. Start with purpose, architecture, stack and entry points, map the directory structure, trace one representative flow end to end, then identify the core abstractions and where changes go. Use when landing in a repository you have never seen and need orientation before making a change.
npx skills add hermeticormus/code-comprehension-skills --skill explore-code
This repository is archived — consider an actively maintained alternative.
Website exploration for testing using Playwright MCP
14.7K installsProfile and explore a dataset to understand its shape, quality, and patterns. Use when encounte…
5.4K installs飞书/Lark 原生 OpenAPI 探索：从官方文档库中挖掘未经 CLI 封?
669.1K installs飞书/Lark 原生 OpenAPI 探索：从官方文档库中挖掘未经 CLI 封?
430.1K installsRelated neighbors and high-traction skills in the same topics — useful to compare before installing.
Rigor Explore compatible skill slug for meaningful and potentially novel deep learning research…
311.2K installsRigor Improve implementation leaf skill for auditable candidate implementation in deep learning…
310.6K installsRigor Improve / Rigor Explore run leaf skill for bounded exploratory evidence in deep learning …
310.4K installsRead-only repository explorer for cold-start orientation, broad cross-file localization, or whe…
39.4K installsGenerate an explore-prompt.md file that gives Momentic's explore agent (`momentic ai explore di…
14.7K installsToken-optimized structural code search using tree-sitter AST parsing. Use instead of reading fu…
6.6K installsDeclared targets from SKILL.md / docs. Unmarked agents are not listed — the skill may still install via the CLI.
Alternate registries and mirrors of this skill.
npx skills add lllllllama/rigorpilot-skills --skill explore-code
npx skills add smithery/brujoh88 --skill explore-code
main
Parsed from SKILL.md frontmatter.
Files included with this skill beyond the listing page.
SKILL.md
2,606 B
SUMMARY.md
344 B
A protocol for building a working mental model of an unfamiliar codebase before changing it. The goal is orientation, not a full audit: enough understanding to know where execution starts, how data moves, and where a given change belongs.
Tradeoff: bias toward building a map before editing. For a one-line fix in a file you already know, skip it.
Before reading source, answer four questions from the README and the manifest:
main, a server listen, a CLI registration, a framework convention).Get the tree, ignoring generated and vendored directories.
tree -L 2 -I 'node_modules|dist|build|.git|target|__pycache__|.venv'Locate source versus tests, config and environment templates, docs, and build and deploy definitions. The layout encodes the team's mental model.
Follow one request, one command, or one public call through every layer: route to handler to service to data access to response, or argument parsing to side effect. One traced flow teaches more than many files read in isolation, because it reveals the layering and the conventions repeated everywhere else.
Name the data models, the seams between layers, the cross-cutting concerns, and the files most changes touch. Then answer the orientation questions:
When you can answer these, you can change the right place instead of the first place that compiles. Produce a short orientation note (the four framing answers, the one flow traced, the core abstractions, and open questions for maintainers) as the artifact.
See full content at https://github.com/HermeticOrmus/code-comprehension-skills.