Skip to content

YurunChen/repo-docs-skills / docs

repo-docs

Flagged

Build and maintain a Markdown guide that helps humans understand a repository through real behavior, concepts, and evidence. Use when a user asks to understand a repo, generate or update repo-docs, answer repo-architecture/onboarding questions, seed docs for a new project, sync docs after code changes, or delete generated repo docs.

FileKindSize
.nojekyllother0 B
EXAMPLES.mdmarkdown33.1 KB
PAGE_RULES.mdmarkdown44.2 KB
QUALITY_RULES.mdmarkdown4.8 KB
README.mdmarkdown8.9 KB
README_CN.mdmarkdown8.6 KB
REFERENCE.mdmarkdown1.0 KB
ROOT_AGENT_RULES.mdmarkdown4.2 KB
SCOPE_MODES.mdmarkdown4.9 KB
SKILL.mdskill md12.4 KB
SYNC_RULES.mdmarkdown20.5 KB
WRITING.mdmarkdown15.0 KB
assets/gsap.min.jsscript70.7 KB
assets/logo.pngbinary284.0 KB
assets/main.pngbinary1.9 MB
assets/repo-docs-evidence-atlas.pngbinary2.1 MB
assets/repo-docs-pixel-kingdom.pngbinary2.6 MB
assets/repo-docs-pixel-workshop.pngbinary2.4 MB
assets/site.cssother63.7 KB
assets/site.jsscript38.6 KB
case-studies/aider.mdmarkdown1.2 KB
case-studies/bolt-diy.mdmarkdown1.2 KB
case-studies/generated/aider/repo-docs/README.mdmarkdown3.7 KB
case-studies/generated/aider/repo-docs/change-log.mdmarkdown1.4 KB
case-studies/generated/aider/repo-docs/code-map.mdmarkdown3.5 KB
case-studies/generated/aider/repo-docs/glossary.mdmarkdown3.3 KB
case-studies/generated/aider/repo-docs/modules/edit-lifecycle.mdmarkdown3.1 KB
case-studies/generated/aider/repo-docs/modules/session-assembly.mdmarkdown2.3 KB
case-studies/generated/aider/repo-docs/references/quality-review.mdmarkdown3.7 KB
case-studies/generated/aider/repo-docs/references/source-evidence.mdmarkdown13.0 KB
case-studies/generated/aider/repo-docs/walkthroughs/one-real-run.mdmarkdown10.4 KB
case-studies/generated/bolt-diy/repo-docs/README.mdmarkdown3.9 KB
case-studies/generated/bolt-diy/repo-docs/change-log.mdmarkdown1.5 KB
case-studies/generated/bolt-diy/repo-docs/code-map.mdmarkdown4.1 KB
case-studies/generated/bolt-diy/repo-docs/glossary.mdmarkdown3.5 KB
case-studies/generated/bolt-diy/repo-docs/modules/chat-api-contract.mdmarkdown5.5 KB
case-studies/generated/bolt-diy/repo-docs/modules/chat-stream.mdmarkdown2.7 KB
case-studies/generated/bolt-diy/repo-docs/modules/provider-model-selection.mdmarkdown2.8 KB
case-studies/generated/bolt-diy/repo-docs/references/quality-review.mdmarkdown3.2 KB
case-studies/generated/bolt-diy/repo-docs/references/source-evidence.mdmarkdown11.1 KB
case-studies/generated/bolt-diy/repo-docs/walkthroughs/one-real-run.mdmarkdown8.6 KB
case-studies/generated/tabby/repo-docs/README.mdmarkdown2.6 KB
case-studies/generated/tabby/repo-docs/change-log.mdmarkdown1.7 KB
case-studies/generated/tabby/repo-docs/code-map.mdmarkdown4.4 KB
case-studies/generated/tabby/repo-docs/glossary.mdmarkdown4.3 KB
case-studies/generated/tabby/repo-docs/modules/completion-contract.mdmarkdown9.6 KB
case-studies/generated/tabby/repo-docs/modules/completion-pipeline.mdmarkdown2.3 KB
case-studies/generated/tabby/repo-docs/modules/retrieval-context.mdmarkdown3.2 KB
case-studies/generated/tabby/repo-docs/references/quality-review.mdmarkdown3.2 KB
case-studies/generated/tabby/repo-docs/references/source-evidence.mdmarkdown10.2 KB
case-studies/generated/tabby/repo-docs/walkthroughs/one-real-run.mdmarkdown9.6 KB
case-studies/tabby.mdmarkdown1.2 KB
index.htmlother26.1 KB
scripts/validate_repo_docs.pyexecutablescript46.6 KB
validate_repo_docs.pyscript482 B

case-studies/generated/aider/repo-docs/references/source-evidence.md

View on GitHub
# Source Evidence

This page is the audit base for the guide. It tracks the inspected source, tests, commands, and caveats behind the CLI edit workflow.

## Evidence Traversal Log

| Pass | Purpose | Inspected evidence | What changed in the model |
| --- | --- | --- | --- |
| Pass 1 | Map the main edit path from terminal input to changed files. | [`pyproject.toml`](https://github.com/Aider-AI/aider/blob/5dc9490bb35f9729ef2c95d00a19ccd30c26339c/pyproject.toml), [`aider/__main__.py`](https://github.com/Aider-AI/aider/blob/5dc9490bb35f9729ef2c95d00a19ccd30c26339c/aider/__main__.py), [`aider/main.py`](https://github.com/Aider-AI/aider/blob/5dc9490bb35f9729ef2c95d00a19ccd30c26339c/aider/main.py), [`aider/coders/base_coder.py`](https://github.com/Aider-AI/aider/blob/5dc9490bb35f9729ef2c95d00a19ccd30c26339c/aider/coders/base_coder.py), [`aider/coders/editblock_coder.py`](https://github.com/Aider-AI/aider/blob/5dc9490bb35f9729ef2c95d00a19ccd30c26339c/aider/coders/editblock_coder.py) | The initial model became: CLI parsing and config produce a `Coder`; `Coder.run()` sends a user request; `send_message()` receives model output; `apply_updates()` parses, checks, writes, and reflects errors. |
| Pass 2 | Challenge the path for boundaries, failure modes, and tests. | [`tests/basic/test_main.py`](https://github.com/Aider-AI/aider/blob/5dc9490bb35f9729ef2c95d00a19ccd30c26339c/tests/basic/test_main.py), [`tests/basic/test_coder.py`](https://github.com/Aider-AI/aider/blob/5dc9490bb35f9729ef2c95d00a19ccd30c26339c/tests/basic/test_coder.py), [`tests/basic/test_editblock.py`](https://github.com/Aider-AI/aider/blob/5dc9490bb35f9729ef2c95d00a19ccd30c26339c/tests/basic/test_editblock.py), dirty-file and dry-run branches in [`aider/coders/base_coder.py`](https://github.com/Aider-AI/aider/blob/5dc9490bb35f9729ef2c95d00a19ccd30c26339c/aider/coders/base_coder.py), failed SEARCH/REPLACE handling in [`aider/coders/editblock_coder.py`](https://github.com/Aider-AI/aider/blob/5dc9490bb35f9729ef2c95d00a19ccd30c26339c/aider/coders/editblock_coder.py) | The model added the non-trivial part: Aider separates model output from file writes, asks before broadening edit scope, protects dirty work with pre-edit commits, reflects malformed edits, and can run lint/test feedback after edits. |

## Understanding Brief

| Field | Guide answer |
| --- | --- |
| Scenario | A developer starts `aider` in a git repo, gives a coding request, and the default diff-style coder applies a SEARCH/REPLACE edit to a file in the chat. |
| Input | The observable input is CLI arguments and a user message. The package exposes the `aider` command through `aider.main:main` in [`pyproject.toml`](https://github.com/Aider-AI/aider/blob/5dc9490bb35f9729ef2c95d00a19ccd30c26339c/pyproject.toml), and `main()` accepts `argv` before parsing config and environment in [`aider/main.py`](https://github.com/Aider-AI/aider/blob/5dc9490bb35f9729ef2c95d00a19ccd30c26339c/aider/main.py). |
| Success / output | A file is changed, and when git auto-commit is enabled the edited file can be committed with an Aider-generated message. Tests assert changed file content and commit counts in [`tests/basic/test_coder.py`](https://github.com/Aider-AI/aider/blob/5dc9490bb35f9729ef2c95d00a19ccd30c26339c/tests/basic/test_coder.py). |
| Hard part | The path is not just "send prompt, write response." It must preserve the user-controlled file set, parse model output into an edit format, avoid clobbering ignored or unapproved files, preserve dirty work for undo, and turn lint/test failures back into model-visible feedback. |
| Boundary / failure | A malformed or non-matching SEARCH block raises a `ValueError`; `apply_updates()` records it as a malformed response and stores a reflected message instead of pretending the edit succeeded. |
| Falsifying check | If `tests/basic/test_editblock.py::TestEditBlockCoder::test_full_edit` stopped showing content changed from `one\ntwo\nthree\n` to `one\nnew\nthree\n`, or if `test_full_edit_dry_run` changed the file during dry run, this walkthrough would be wrong. |
| Likely reader follow-up | A careful reader will ask how Aider chooses the exact edit format or model. This first guide answers only the default edit lifecycle and points to source evidence for model/config selection; deeper model registry behavior is out of scope. |

## Coverage Notes

Covered: the terminal CLI startup path, `Coder` selection, a default diff edit, edit permission checks, dirty-file pre-commit, malformed edit reflection, lint/test reflection hooks, and auto-commit.

Checked but not traced end to end: GUI launch, watch mode, copy/paste mode, repo-map ranking internals, model metadata registration, OAuth onboarding, voice input, browser scraping, architect/editor split, and help search. These are adjacent surfaces that reuse or feed the `Coder` loop but deserve their own walkthroughs.

## Claim Audit

| Claim | Evidence | Confidence | Caveat | Used by |
| --- | --- | --- | --- | --- |
| The installed CLI command enters `aider.main:main`. | [`pyproject.toml`](https://github.com/Aider-AI/aider/blob/5dc9490bb35f9729ef2c95d00a19ccd30c26339c/pyproject.toml) declares `aider = "aider.main:main"`, and [`aider/__main__.py`](https://github.com/Aider-AI/aider/blob/5dc9490bb35f9729ef2c95d00a19ccd30c26339c/aider/__main__.py) imports and calls `main()`. | Confirmed | Package installation behavior was inspected from project metadata, not exercised by installing the package. | [`README.md`](../README.md), [`walkthroughs/one-real-run.md`](../walkthroughs/one-real-run.md) |
| `main()` resolves git/config context before creating a coder. | [`aider/main.py`](https://github.com/Aider-AI/aider/blob/5dc9490bb35f9729ef2c95d00a19ccd30c26339c/aider/main.py) starts with git root detection, config search, parser construction, dotenv loading, and argument reparsing. | Confirmed | The guide does not document every CLI option or config precedence edge. | [`code-map.md`](../code-map.md), [`modules/session-assembly.md`](../modules/session-assembly.md) |
| `main()` assembles the objects that drive an edit session and passes them into `Coder.create()`. | [`aider/main.py`](https://github.com/Aider-AI/aider/blob/5dc9490bb35f9729ef2c95d00a19ccd30c26339c/aider/main.py) creates `Commands`, `ChatSummary`, map token settings, and calls `Coder.create()` with repo, files, lint/test, commit, map, stream, and language options. | Confirmed | Some objects have deeper behavior outside this guide, especially model registry and repo-map internals. | [`walkthroughs/one-real-run.md`](../walkthroughs/one-real-run.md), [`code-map.md`](../code-map.md), [`modules/session-assembly.md`](../modules/session-assembly.md) |
| `Coder.create()` chooses the concrete coder class by edit format. | [`aider/coders/base_coder.py`](https://github.com/Aider-AI/aider/blob/5dc9490bb35f9729ef2c95d00a19ccd30c26339c/aider/coders/base_coder.py) resolves `edit_format` from the caller, previous coder, or model default, then scans `aider.coders.__all__` for a matching class. | Confirmed | The exact set of available coder classes is not listed in this guide. | [`walkthroughs/one-real-run.md`](../walkthroughs/one-real-run.md) |
| The default diff-style coder parses SEARCH/REPLACE blocks from model output. | [`aider/coders/editblock_coder.py`](https://github.com/Aider-AI/aider/blob/5dc9490bb35f9729ef2c95d00a19ccd30c26339c/aider/coders/editblock_coder.py) has `EditBlockCoder.edit_format = "diff"` and `get_edits()` delegates to `find_original_update_blocks()`. | Confirmed | "Default" depends on model settings; this guide follows the diff path because tests and source make it concrete. | [`walkthroughs/one-real-run.md`](../walkthroughs/one-real-run.md), [`code-map.md`](../code-map.md), [`modules/edit-lifecycle.md`](../modules/edit-lifecycle.md) |
| `send_message()` owns the model-call loop, retries, output-limit handling, response capture, and post-response edit phase. | [`aider/coders/base_coder.py`](https://github.com/Aider-AI/aider/blob/5dc9490bb35f9729ef2c95d00a19ccd30c26339c/aider/coders/base_coder.py) appends the user message, formats messages, checks tokens, sends with retry handling, captures `partial_response_content`, then calls `apply_updates()`. | Confirmed | Actual provider calls are not exercised in guide verification; tests stub `coder.send`. | [`walkthroughs/one-real-run.md`](../walkthroughs/one-real-run.md) |
| File writes are gated by a dry-run parse before permission checks and final writes. | [`aider/coders/base_coder.py`](https://github.com/Aider-AI/aider/blob/5dc9490bb35f9729ef2c95d00a19ccd30c26339c/aider/coders/base_coder.py) calls `get_edits()`, `apply_edits_dry_run()`, `prepare_to_edit()`, and then `apply_edits()`. | Confirmed | Each coder class may implement `apply_edits_dry_run()` differently. | [`modules/edit-lifecycle.md`](../modules/edit-lifecycle.md) |
| Aider asks before creating or editing files outside the current chat set, and skips gitignored files. | [`aider/coders/base_coder.py`](https://github.com/Aider-AI/aider/blob/5dc9490bb35f9729ef2c95d00a19ccd30c26339c/aider/coders/base_coder.py) implements `allowed_to_edit()` with checks for in-chat files, gitignore, new-file confirmation, and confirmation for files not added to chat. | Confirmed | Tests in this pass focused on edit/commit behavior, not every prompt branch. | [`modules/edit-lifecycle.md`](../modules/edit-lifecycle.md) |
| Dirty files targeted by model edits can be committed before the edit, preserving an undoable baseline. | [`aider/coders/base_coder.py`](https://github.com/Aider-AI/aider/blob/5dc9490bb35f9729ef2c95d00a19ccd30c26339c/aider/coders/base_coder.py) marks dirty paths in `check_for_dirty_commit()` and commits them in `dirty_commit()`; [`tests/basic/test_coder.py`](https://github.com/Aider-AI/aider/blob/5dc9490bb35f9729ef2c95d00a19ccd30c26339c/tests/basic/test_coder.py) asserts three commits when editing a dirty file. | Confirmed | Requires git and `dirty_commits` behavior; no-git mode has different persistence behavior. | [`walkthroughs/one-real-run.md`](../walkthroughs/one-real-run.md), [`code-map.md`](../code-map.md), [`modules/edit-lifecycle.md`](../modules/edit-lifecycle.md) |
| Aider commits only edited files for an Aider edit, not unrelated dirty files. | [`tests/basic/test_coder.py`](https://github.com/Aider-AI/aider/blob/5dc9490bb35f9729ef2c95d00a19ccd30c26339c/tests/basic/test_coder.py) keeps `file1.txt` dirty while editing `file2.txt`, asserts dirty content is absent from generated commit-message diffs, and leaves `file1.txt` dirty. | Confirmed | The test stubs commit-message generation but verifies the diff scope. | [`modules/edit-lifecycle.md`](../modules/edit-lifecycle.md) |
| Non-matching SEARCH blocks become reflected feedback instead of silent success. | [`aider/coders/editblock_coder.py`](https://github.com/Aider-AI/aider/blob/5dc9490bb35f9729ef2c95d00a19ccd30c26339c/aider/coders/editblock_coder.py) builds a `SearchReplaceNoExactMatch` message and raises `ValueError`; [`aider/coders/base_coder.py`](https://github.com/Aider-AI/aider/blob/5dc9490bb35f9729ef2c95d00a19ccd30c26339c/aider/coders/base_coder.py) catches `ValueError`, increments malformed response count, and stores `reflected_message`. | Confirmed | The guide names the mechanism, not every exact error string variant. | [`walkthroughs/one-real-run.md`](../walkthroughs/one-real-run.md), [`modules/edit-lifecycle.md`](../modules/edit-lifecycle.md) |
| After edits, Aider may auto-commit, run lint, reflect lint errors, run shell commands, and run tests. | [`aider/coders/base_coder.py`](https://github.com/Aider-AI/aider/blob/5dc9490bb35f9729ef2c95d00a19ccd30c26339c/aider/coders/base_coder.py) calls `auto_commit()`, `lint_edited()`, `run_shell_commands()`, and `cmd_test()` after `apply_updates()`. | Confirmed | The guide does not trace command execution internals. | [`walkthroughs/one-real-run.md`](../walkthroughs/one-real-run.md) |
| `--apply` reuses the edit application path without sending a model request. | [`aider/main.py`](https://github.com/Aider-AI/aider/blob/5dc9490bb35f9729ef2c95d00a19ccd30c26339c/aider/main.py) reads an apply file into `coder.partial_response_content` and calls `coder.apply_updates()`. | Confirmed | Mentioned as a verification shortcut, not as the main behavior. | [`walkthroughs/one-real-run.md`](../walkthroughs/one-real-run.md) |

## Verification Commands

The guide was checked against source inspection and is intended to be verified with:

```bash
python -m pytest tests/basic/test_editblock.py::TestUtils::test_full_edit tests/basic/test_editblock.py::TestUtils::test_full_edit_dry_run tests/basic/test_coder.py::TestCoder::test_gpt_edit_to_dirty_file tests/basic/test_coder.py::TestCoder::test_only_commit_gpt_edited_file
python path/to/repo-docs-skill/scripts/validate_repo_docs.py repo-docs --repo-root .
```

Current verification caveat: the focused pytest command was attempted on 2026-06-29 13:30 CST, but this Python environment is missing `oslex`, so pytest failed during import before executing the selected tests. `oslex==0.1.3` is listed in [`requirements.txt`](https://github.com/Aider-AI/aider/blob/5dc9490bb35f9729ef2c95d00a19ccd30c26339c/requirements.txt).

Evidence status: Confirmed unless noted.
rea-changelog-updatemorluto/reaPrepare or rewrite REA release changelogs and GitHub release notes from pinned Git history, with verified contributor thanks and Release Please synchronization.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
academy-guideanthropics/skillsStop and check this skill before finishing any reply to a question about how to use Claude or a Claude product — it recommends matching courses, tutorials, and use cases from Claude Academy (academy.claude.com), Anthropic's learning hub. Trigger on: "how do I", "how can I", "getting started with", "what can Claude do", "teach me", "learn to use"; questions about artifacts, projects, skills, plugins, connectors, MCP; requests about rolling Claude out to a team, class, or organization; and any ask for training materials, onboarding content, or learning resources. Use it when the user is learning how to use a feature or product — not when they are mid-task and just want the task done. This skill composes with other skills: after consulting product documentation to answer how a Claude feature works, also check here for a matching course or tutorial — a docs-grounded answer and an Academy recommendation belong together. Only recommend on a strong match; never invent Academy content.Stars: 198kUpdated Claude CodeCodexVerifiedInstall
claude-apianthropics/skillsReference for the Claude API / Anthropic SDK — model ids, pricing, params, streaming, tool use, MCP, agents, caching, token counting, model migration. TRIGGER — read BEFORE opening the target file; don't skip because it "looks like a one-liner" — whenever: the prompt names Claude/Anthropic in any form (Claude, Anthropic, Fable, Opus, Sonnet, Haiku, `anthropic`, `@anthropic-ai`, `claude-*`, `us.anthropic.*`, `[1m]`); the user asks about an LLM (pricing/model choice/limits/caching) — never answer from memory; OR the task is LLM-shaped with provider unstated (agent/MCP/tool-definition/multi-agent/RAG/LLM-judge/computer-use; generate/summarize/extract/classify/rewrite/converse over NL; debugging refusals/cutoffs/streaming/tool-calls/tokens). SKIP only when another provider is being worked on (overrides all triggers): OpenAI/GPT/Gemini/Llama/Mistral/Cohere/Ollama named in the query; OR `grep -rE 'openai|langchain_openai|google.generativeai|genai|mistralai|cohere|ollama'` over the project hits (run this grep FIRST if no provider named — don't Read the file).Stars: 198kUpdated Claude CodeCodexUnverifiedInstall
doc-coauthoringanthropics/skillsGuide users through a structured workflow for co-authoring documentation. Use when user wants to write documentation, proposals, technical specs, decision docs, or similar structured content. This workflow helps users efficiently transfer context, refine content through iteration, and verify the doc works for readers. Trigger when user mentions writing docs, creating proposals, drafting specs, or similar documentation tasks.Stars: 198kUpdated Claude CodeCodexVerifiedInstall