New: 97,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Codexqa Change Analysis

by openqa-cn in openqa-cn/codexqa

Change-impact analysis for one diff: use CodexQA diff indexing (index --diff-base tags nodes and files add / change / delete, then change-groups / symbol-diff / file-base) to write one HTML report…

Apache-2.0Auto-check passedTesting & QA

Install Codexqa Change Analysis

skills CLI
$ npx skills add openqa-cn/codexqa --skill codexqa-change-analysis -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install openqa-cn/codexqa codexqa-change-analysis --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/openqa-cn/codexqa.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/codexqa-change-analysis .claude/skills/codexqa-change-analysis && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
codexqa-change-analysis
GitHub stars
152
Token cost
~3.1k tokens
SKILL.md length
1,449 words
Files
10 (incl. references, assets)
Skills in repo
14
Repo updated
First seen
Licence
Apache-2.0

At a glance

Change-impact analysis for one diff: use CodexQA diff indexing (index --diff-base tags nodes and files add / change / delete, then change-groups / symbol-diff / file-base) to write one HTML report…

  • Works in 9 steps: Do not conclude without a trusted diff… → "An entry is a method with no upstream… → Test coverage comes only from a tests… → …
  • The user mentions change impact
  • SKILL.md covers Deliverables (both required), Hard rules (a violation means…, Why git / grep are still used and Index trust gate (do this first), plus 3 more sections
  • Calls git

What it does

Codexqa Change Analysis is an agent skill from openqa-cn/codexqa. Change-impact analysis for one diff: use CodexQA diff indexing (index --diff-base tags nodes and files add / change / delete, then change-groups / symbol-diff / file-base) to write one HTML report (affected entries, changed methods, test plan, coverage verdict, sensitive paths) and add runnable test files in the analyzed repo for uncovered points (add only; do not edit existing test files). Use when the user mentions change impact, change analysis, blast radius, affected entries, added methods, changed methods…

Its SKILL.md is about 3.1k tokens, which your agent loads only when the skill is triggered. The skill folder holds 11 other files, including reference files and assets (for example `KNOWN_LIMITATIONS.md`, `KNOWN_LIMITATIONS.zh-CN.md` and `README.md`). Compatibility notes: Requires Node.js = 18 and the codexqa CLI (npm i -g @openqa-cn/codexqa). Diff indexing and queries need no LLM. Data lives in ~/.codexqa/. Without codexqa…

It sits in Testing & QA, covering Test generation. The repository describes itself as: codexqa: 11 local-first Agent Skills for Cursor, Claude Code, Codex & OpenClaw — change impact analysis, AI code review, defect scan, testcase generation, browser replay & RCA. The licence is Apache-2.0.

When your agent uses it

  • The user mentions change impact
  • Change analysis
  • Affected entries
  • Changed methods

Example prompts

  • “/codexqa-change-analysis”

Requirements

  • Node.js
  • Compatibility (from SKILL.md): Requires Node.js >= 18 and the `codexqa` CLI (`npm i -g @openqa-cn/codexqa`). Diff indexing and queries need no LLM. Data lives in ~/.codexqa/. Without codexqa, follow Degraded mode. Generating cases requires the analyzed repo to start; the end-to-end layer also needs a browser (L1/L2 in references/generate-cases.md).

Workflow steps

9 steps, taken from the first numbered list in SKILL.md.

  1. Do not conclude without a trusted diff index. --diff-base must be paired with --full. Re-run whenever the base changes (incremental mode…
  2. "An entry is a method with no upstream caller" cannot rest on to_count == 0 alone. Cross-file call edges are missing in dynamic-language…
  3. Test coverage comes only from a tests edge or a real assertion in a test file. A filename containing *test*, or a symbol with the same…
  4. Every conclusion must be checkable: command, node id, and file:line. Do not replace query results with README, docs, or name guesses.
  5. Report only changes introduced by this diff. Pre-existing issues found along the way go in their own note, not in the change list.
  6. Do not enumerate call sites with codexqa query search. It is file-chunk BM25: one hit per file, no call line. Use edges / reach when the…
  7. The report is a result for a reviewer, not an analysis log. Process data (true-entry criteria, missing-edge fix counts, stub ratios, query…
  8. Line numbers in the report are editor lines (1-based). CodexQA start_line / end_line / search lines are 0-based. grep -n and editors are…
  9. Generated cases are add-only. Do not edit existing test files (and do not edit package.json). Write the missing assertion as a new case…

What it can do on your machine

Read from SKILL.md and the folder at commit 7839542. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • git

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md. Its commands use git, which can reach the network depending on how they are called.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

  • Compatibility

    Requires Node.js >= 18 and the `codexqa` CLI (`npm i -g @openqa-cn/codexqa`). Diff indexing and queries need no LLM. Data lives in ~/.codexqa/. Without codexqa, follow Degraded mode. Generating cases requires the analyzed repo to start; the end-to-end layer also needs a browser (L1/L2 in references/generate-cases.md).

    From compatibility in the SKILL.md frontmatter.

Context cost

Codexqa Change Analysis loads about 3.1k tokens when it runs, and up to ~19k if it reads all its reference files. Until then it costs about 215 tokens; SKILL.md has 1,449 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~215
When it runs · the whole SKILL.md, loaded when a task matches
~3.1k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~19k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from openqa-cn/codexqa at commit 7839542, republished under its Apache-2.0 licence (© openqa-cn). 1,449 words, ~3,083 tokens.

Download SKILL.mdSave it as .claude/skills/codexqa-change-analysis/SKILL.md (or your agent's skills folder). This skill also uses 9 other files; get the full folder from GitHub.
name
codexqa-change-analysis
description
Change-impact analysis for one diff: use CodexQA diff indexing (`index --diff-base` tags nodes and files `add` / `change` / `delete`, then `change-groups` / `symbol-diff` / `file-base`) to write one HTML report (affected entries, changed methods, test plan, coverage verdict, sensitive paths) and add runnable test files in the analyzed repo for uncovered points (add only; do not edit existing test files). Use when the user mentions change impact, change analysis, blast radius, affected entries, added methods, changed methods, test plan, generate or recall tests, link existing tests, entry points, 变更分析, 变更代码分析, 影响面, 影响入口, 测试方案, or 召回测试用例. Requires the codexqa CLI. One change only. Whole-repo analysis with no diff, and general symbol-graph Q&A, are codexqa-code-analyzer, not this skill. Former skill name: change-impact-analysis.
compatibility
Requires Node.js >= 18 and the `codexqa` CLI (`npm i -g @openqa-cn/codexqa`). Diff indexing and queries need no LLM. Data lives in ~/.codexqa/. Without codexqa, follow Degraded mode. Generating cases requires the analyzed repo to start; the end-to-end layer also needs a browser (L1/L2 in references/generate-cases.md).
license
Apache-2.0
metadata.author
openqa-cn
metadata.version
1.0.0
metadata.open-standard
agentskills

Change Impact Analysis

One sentence: build a trusted diff index, then deliver two artifacts — a 7-section HTML report (key findings / risks and unknowns / affected entries / change list / test plan / coverage verdict / sensitive paths) plus runnable test files for points that are not covered. codexqa has no diff or review subcommand. Change analysis always follows this path:

text
index --diff-base <base> --full  →  change-groups / files --change  →  symbol-diff
   →  reach --direction in (entries)  →  tested_count / reach --edge-kinds tests  →  test-file text recall

Deliverables (both required)

IDDeliverableForm
AHTML report: key findings / risks and unknowns / affected entries / change list / test plan / coverage verdict / sensitive pathsOne self-contained HTML file at the repo root: change-impact-<repo-slug>-YYYYMMDD-HHMM.html
BNew test files: turn every test-plan row whose disposition is "add" into a runnable caseNew files in the repo's existing test tree (at least one API/integration file; add an end-to-end file when a browser is required). Add only; do not edit existing test files. Layer rules, names, and fallbacks: references/generate-cases.md

Recalled existing cases are not a separate deliverable. They land in the test-plan table's "case file" column and in the coverage-verdict section.

Template: assets/report-template.html (shipped with this skill). Fill contract: references/report.md — read it before writing the report. cp the template and edit by id. Do not read the whole template into context (the 23 ids and the structure are in report.md).

Default language is English. The report is not Markdown in the chat; the chat only returns the HTML path. Copy the template, then Edit the slots. Do not change CSS. Do not invent a new layout. Delete the trailing mermaid loader as step 5 of "Diagrams must be static" in references/report.md. Do not touch it otherwise.

Section order is fixed: results first, process last. The body answers only the four questions a reviewer cares about (which methods changed / which APIs or pages can trigger them / what to test / which existing cases can be reused). Method and process data never enter the report (no process appendix, no "confidence / analysis scope" section). They stay in the analysis.

Hard rules (a violation means rewrite)

  1. Do not conclude without a trusted diff index. --diff-base must be paired with --full. Re-run whenever the base changes (incremental mode does not refresh change tags; see KNOWN_LIMITATIONS.md item 1).
  2. "An entry is a method with no upstream caller" cannot rest on to_count == 0 alone. Cross-file call edges are missing in dynamic-language repos. Cross-check with text search / imports, or a function called by router.get('/api/...') will be reported as an entry.
  3. Test coverage comes only from a tests edge or a real assertion in a test file. A filename containing *test*, or a symbol with the same name, is not coverage.
  4. Every conclusion must be checkable: command, node id, and file:line. Do not replace query results with README, docs, or name guesses.
  5. Report only changes introduced by this diff. Pre-existing issues found along the way go in their own note, not in the change list.
  6. Do not enumerate call sites with codexqa query search. It is file-chunk BM25: one hit per file, no call line. Use edges / reach when the graph has the edge, and grep when it does not. See the split below.
  7. The report is a result for a reviewer, not an analysis log. Process data (true-entry criteria, missing-edge fix counts, stub ratios, query fields, command names) never enters the report (no process appendix, and no confidence paragraph under "Risks and unknowns"). Headline numbers, key findings, and the entry table describe only the change's impact and risk, in business / API language. See "Iron rule" in references/report.md.
  8. Line numbers in the report are editor lines (1-based). CodexQA start_line / end_line / search lines are 0-based. grep -n and editors are 1-based. Add 1 to every CodexQA line before writing file:line, or the reader lands one line off (measured on one JS repo: all 11 method locations were off by 1). See KNOWN_LIMITATIONS.md item 8.
  9. Generated cases are add-only. Do not edit existing test files (and do not edit package.json). Write the missing assertion as a new case; do not edit the old one. "Add only" does not cap the file count — if a point needs a browser, create a new end-to-end file; do not mark it "cannot cover". Layer rules, file names (follow the repo language and existing test style), and fallbacks are in references/generate-cases.md. You must actually start the service and run the files, and write pass / fail counts back into the report. A case that was never run was not delivered. See references/generate-cases.md.

Why git / grep are still used

Change detection always belongs to codexqa (index --diff-base --full → change-groups / files --change). git and grep fill only measured CodexQA gaps. They do not bypass CodexQA:

ToolRole in this skillWhy CodexQA alone is not enough
codexqa index/queryChange detection, patches, edge walks, blast radius, tests edges — primary path—
git rev-parse / git diff --name-onlyResolve the base and branch; check whether CodexQA change tags are staleIncremental indexes keep the previous diff tags (measured; KNOWN_LIMITATIONS.md item 1)
grep -rnEnumerate call sites missing from the graph, and locate route registration / main and other text entriesWhen edges are missing, edges / reach return nothing; search is file-level and misses the real call (items 2 and 6)
git diff (degraded mode)Fallback when CodexQA or the index is unavailableOnly then

Rule: if the graph can answer it, use the graph (who is affected, where the entry is, whether a tests edge exists). Fall back to text search only when the graph is clearly missing an edge, or there is no CodexQA index. Whether evidence came from the graph or from text is a criterion; do not write it into the report. When the whole report is degraded, mark #meta-line with one phrase: "Degraded: graph incomplete".

Show full SKILL.md (504 more words)Show less

Index trust gate (do this first)

bash
cd <repo>                                   # prefer a local checkout
git rev-parse --abbrev-ref HEAD             # current branch
git diff --name-only <base>...HEAD          # files git considers changed (the reference set)

codexqa index . --diff-base <base> --full   # --full is required
codexqa query --repo . change-groups        # empty groups = the index is not in diff mode
codexqa query --repo . files --change add,change,delete

Choose <base> in this order: user-specified → origin/main → main / master → HEAD~1. If there are several branches and the base is unclear, ask the user. Do not guess.

Consistency check: the git diff --name-only set must match files --change add,change,delete. If they differ, re-run index --diff-base <base> --full. If they still differ, mark #meta-line "Degraded: index untrusted" and degrade.

Workflow (8 steps)

StepWhatDetail
1Change list: added / changed / deleted methods and filesreferences/analysis.md §1
2Read the patch; classify the change (signature / behavior / add / delete)§2
3Entry resolution: walk up to methods with no upstream caller, then force a cross-check§3
4Regression scope: direct and transitive callers§4
5Test plan: case matrix by entry, plus sensitive paths§5
6Recall existing tests: layer A tests edges → layer B text recall (results go into the test plan and are written in step 8). On a re-run, move the previous change-impact-* artifacts out of the repo first, or recall will treat them as existing casesreferences/test-recall.md
7Generate case code: turn rows whose disposition is "add" into runnable cases (file count and names follow the layer rules). Run them and record file:line plus pass / skip countsreferences/generate-cases.md
8Write the report: copy the HTML template → fill by id (including generated-case results) → tell the user the pathreferences/report.md

Read references/analysis.md first. Read references/test-recall.md only at step 6. Recalled cases go into the test-plan "case file" column and the coverage-verdict section. Do not add a separate recall section. Step 7 must finish before the report is written. The file:line of rows marked "add", the "now covered" lines in #coverage, and the pass / skip counts in #testplan-artifact all come from a real run of the generated files. Writing the report first means filling those slots twice. Read references/generate-cases.md at step 7. Read references/report.md before step 8, and use the shipped assets/report-template.html. Do not preload the whole tree.

Field cheat sheet: to_count = in-degree (caller count; used to judge entries); from_count = out-degree (who it calls; not blast radius); tested_count = number of tests edges pointing at it. Commands live next to each step in references/analysis.md. They are not repeated here.

Degraded mode (no CodexQA, or the index is unusable)

Do not drop the delivery. Use git plus text search. The deliverable is still one HTML file. Mark #meta-line with "Degraded: no symbol graph" (the reader needs to know how strong the conclusion is). Do not write the degraded details into the report:

StepDegraded approach
Change listgit diff --name-status <base>...HEAD; use git diff -U0 hunk headers to locate method bounds
Patchgit diff <base>...HEAD -- <file>
Entriesgrep -rn "<symbol>" for every call site; walk those callers upward; route registration / main / an export is an entry
Test recallGo straight to layer B in references/test-recall.md

Rewrite conditions (any one means rewrite the whole report) are in "Rewrite the whole report if" in references/report.md. That list is the complete one; it is not repeated here.

When blocked

Symptom → action is item 10 in KNOWN_LIMITATIONS.md.

© openqa-cn, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 9 other files (references, assets) in skills/codexqa-change-analysis of openqa-cn/codexqa.

  • SKILL.md
  • KNOWN_LIMITATIONS.md
  • KNOWN_LIMITATIONS.zh-CN.md
  • README.md
  • README.zh-CN.md
  • assets/report-template.html
  • references/analysis.md
  • references/generate-cases.md
  • references/report.md
  • references/test-recall.md

Open the folder on GitHubat commit 7839542

Compare with similar skills

Codexqa Change Analysis next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Codexqa Change Analysis compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Codexqa Change Analysis this skillopenqa-cn/codexqa152—~3.1kAutomated safety check: PassApache-2.0
Emcaklofas/kicad-happy1.3k1 repos~2.8kAutomated safety check: PassMIT
Swig Testswig/swig6.3k—~2.3kAutomated safety check: PassCustom licence
Generate Test Cases342164796/generate-test-cases1191 repos~2.9kAutomated safety check: PassNone
Verify Cc Safety Netkenryu42/cc-safety-net1.6k—~2kAutomated safety check: PassMIT
Wioworkersio/skills180—~5.8kAutomated safety check: PassMIT

Similar skills

  • Emc

    aklofas/kicad-happy

    EMC pre-compliance risk analysis for KiCad PCB designs — 18 check categories, 44 rule IDs covering ground planes, decoupling, I/O filtering, switching harmonics, clock routing, differential pair…

    1.3k GitHub starsUsed in 1 repo~2.8k tokens
    Testing & QAAuto-check passed

  • Swig Test

    swig/swig

    Run SWIG test suite for specific languages. An agent skill from swig/swig.

    6.3k GitHub stars~2.3k tokensUpdated today
    Testing & QAAuto-check passed

  • Generate Test Cases

    342164796/generate-test-cases

    自主学习型测试文档生成器。从需求文档（Markdown）生成测试用例 XMind 文件，支持持久化记忆和持续学习。当用户提到"生成测试用例"、"根据需求生成测试"时触发。

    119 GitHub starsUsed in 1 repo~2.9k tokens
    Testing & QAAuto-check passed

  • Verify Cc Safety Net

    kenryu42/cc-safety-net

    Launch and drive the real cc-safety-net CLI — the hook decision path, explain, status/doctor, logs, and the local policy GUI — against an isolated home, capturing evidence.

    1.6k GitHub stars~2k tokensUpdated yesterday
    Testing & QAAuto-check passed

  • Wio

    workersio/skills

    Testing workflow skill for finding high-value test candidates, writing focused tests, generating realistic workloads, reviewing test value, and diagnosing test-suite health.

    180 GitHub stars~5.8k tokensUpdated 2 mo ago
    Testing & QAAuto-check passed

  • File Server

    microsoft/WindowsProtocolTestSuites

    Official

    ALWAYS LOAD THIS SKILL when working with FileServer, SMB, SMB2, SMB3, CIFS, file sharing, MS-SMB2, MS-FSCC, MS-FSA, MS-DFSC, MS-FSRVP, MS-RSVD, MS-SQOS, or any file server protocol test…

    567 GitHub stars~4.1k tokensUpdated 22 days ago
    Testing & QAAuto-check passed

More from openqa-cn/codexqa

All 14 skills in this repo

  • Codexqa Code Wiki

    openqa-cn/codexqa

    Builds a local architecture wiki for a repository from the CodexQA symbol graph (no model needed): modules, who calls whom and how often, reading paths, and one self-contained HTML page.

    152 GitHub stars~1.3k tokensUpdated 4 days ago
    Auto-check passed

  • Codexqa Rootcause Analyzer

    openqa-cn/codexqa

    Diagnoses exception root causes from stack traces, logs, call-chain dumps, and debug output using the CodexQA CLI for structured repo analysis.

    152 GitHub stars~2.6k tokensUpdated 4 days ago
    Auto-check passed

  • Codexqa Testdata Generator

    openqa-cn/codexqa

    Constructs test data against real backends and writes it back into test cases as executable preconditions.

    152 GitHub stars~4.8k tokensUpdated 4 days ago
    Auto-check passed

  • Codexqa Skill Router

    openqa-cn/codexqa

    Auto-routes a user request to the matching codexqa skill, then ensures that skill is on disk and follows its SKILL.md.

    152 GitHub stars~2k tokensUpdated 4 days ago
    Auto-check passed

  • Codexqa Testcase Generator

    openqa-cn/codexqa

    Generates test plans and test cases from local requirements for APP, Web, and server.

    152 GitHub stars~3.6k tokensUpdated 4 days ago
    Auto-check passed

  • Catalog

    openqa-cn/codexqa

    Constructs catalog products (standard or limited), catalog orders, and account-credit enrollment, including product-then-credit scenes.

    152 GitHub stars~514 tokensUpdated 4 days ago
    Auto-check passed

Categories

Questions about Codexqa Change Analysis

What does Codexqa Change Analysis do?

Change-impact analysis for one diff: use CodexQA diff indexing (index --diff-base tags nodes and files add / change / delete, then change-groups / symbol-diff / file-base) to write one HTML report…. Codexqa Change Analysis is an agent skill from openqa-cn/codexqa. Change-impact analysis for one diff: use CodexQA diff indexing (index --diff-base tags nodes and files add / change / delete, then change-groups / symbol-diff / file-base) to write one HTML report (affected entries, changed methods, test plan, coverage verdict, sensitive paths) and add runnable test files in the analyzed repo for uncovered points (add only; do not edit existing test files).

When should I use Codexqa Change Analysis?

Codexqa Change Analysis fits situations like: the user mentions change impact; change analysis; affected entries; changed methods.

How do I install Codexqa Change Analysis in Claude Code?

Run `npx skills add openqa-cn/codexqa --skill codexqa-change-analysis -a claude-code`. Or copy the skill folder (skills/codexqa-change-analysis in openqa-cn/codexqa) into .claude/skills/codexqa-change-analysis in your project. Claude Code loads it when a task matches its description.

How do I install Codexqa Change Analysis in Codex?

Run `npx skills add openqa-cn/codexqa --skill codexqa-change-analysis -a codex`. Or copy the skill folder (skills/codexqa-change-analysis in openqa-cn/codexqa) into .agents/skills/codexqa-change-analysis in your project. Codex loads it when a task matches its description.

Can I use Codexqa Change Analysis in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add openqa-cn/codexqa --skill codexqa-change-analysis -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/codexqa-change-analysis, .gemini/skills/codexqa-change-analysis, .github/skills/codexqa-change-analysis and .opencode/skills/codexqa-change-analysis in your project.

What does Codexqa Change Analysis need to run?

Going by SKILL.md and its folder, Codexqa Change Analysis needs the command-line tools its instructions call (git). Our summary lists: Node.js. Compatibility (from SKILL.md): Requires Node.js >= 18 and the `codexqa` CLI (`npm i -g @openqa-cn/codexqa`). Diff indexing and queries need no LLM. Data lives in ~/.codexqa/. Without codexqa, follow Degraded mode. Generating cases requires the analyzed repo to start; the end-to-end layer also needs a browser (L1/L2 in references/generate-cases.md). .

Does Codexqa Change Analysis access the network?

SKILL.md contains no URLs. Its commands use git, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Codexqa Change Analysis safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Codexqa Change Analysis use?

Codexqa Change Analysis is published under the Apache-2.0 licence (declared in SKILL.md). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Codexqa Change Analysis use?

About 3.1k tokens (SKILL.md is roughly 12k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 16k tokens, read only when the agent opens those files.

What are the alternatives to Codexqa Change Analysis?

Skills that share tags, products or a category with Codexqa Change Analysis: Emc (aklofas/kicad-happy, 1.3k stars), Swig Test (swig/swig, 6.3k stars), Generate Test Cases (342164796/generate-test-cases, 119 stars) and Verify Cc Safety Net (kenryu42/cc-safety-net, 1.6k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Codexqa Change Analysis?

openqa-cn (a GitHub organization) maintains it in openqa-cn/codexqa, which has 152 GitHub stars. The repository holds 14 skills in this directory. The repository was last updated on October 3, 2026.

Source: openqa-cn/codexqa on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.