# AI use cases for Codexqa (openqa-cn/codexqa) Repo: [openqa-cn/codexqa](https://github.com/openqa-cn/codexqa) | Use cases: 5 | Page: https://dedupedai.com/use-cases/openqa-cn-codexqa/ CodexQA is an open-source pack of local Agent Skills for Cursor, Claude Code, Codex, and OpenClaw that verifies code quality, change impact, defects, requirements, test cases, and browser flows before AI-generated code is merged. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Review a branch before merge and get a shareable HTML report: https://dedupedai.com/use-cases/openqa-cn-codexqa/review-branch-before-merge-with-html-report/ - Scan a branch for bugs and security issues before merging: https://dedupedai.com/use-cases/openqa-cn-codexqa/scan-branch-for-bugs-and-security-issues/ - Map a diff's impact and generate tests for uncovered paths: https://dedupedai.com/use-cases/openqa-cn-codexqa/find-untested-paths-from-one-diff/ - Find PRD gaps and conflicts before anyone writes code: https://dedupedai.com/use-cases/openqa-cn-codexqa/find-requirement-gaps-before-coding/ - Replay a browser flow and check the evidence it shows: https://dedupedai.com/use-cases/openqa-cn-codexqa/replay-browser-flow-and-check-visible-evidence/ ## How to review a branch before merge and get a shareable HTML report with Codexqa Repo: [openqa-cn/codexqa](https://github.com/openqa-cn/codexqa) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Produce a bilingual REVIEW-REPORT.html with merge risk, call chains, and test gaps for a reviewer. **Who it is for:** A developer or tech lead who has an AI-generated branch ready and needs a reviewer-ready report before the merge gate. **The problem:** A raw git diff does not show call chains, blast radius, or which tests are missing, so reviewers miss risk. **The outcome:** A REVIEW-REPORT.html on disk that lists merge risk, gaps, and what still needs tests, ready to attach to the review. ### Steps 1. Install the code reviewer skill with `npx skills add openqa-cn/codexqa --skill codexqa-code-reviewer`. 2. Pick your agent (Cursor, Claude Code, Codex, or OpenClaw) when prompted. 3. Check out the branch you want to review in your local repository. 4. Open a new agent session and paste the agent prompt below. 5. Open the generated REVIEW-REPORT.html and attach it to the review request. ### Prompt to paste into your agent ```text Use the codexqa-code-reviewer skill in this repo to review the current branch against origin/main. Collect the CodexQA evidence pack (callers, blast radius, test edges) and write a bilingual REVIEW-REPORT.html. Verify that the report lists merge risk and the test gaps it found. ``` ### Caveat The README says the agent you already use is the runtime, so report quality depends on the model behind that agent, and the README gives no benchmark for the reviewer skill. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/openqa-cn-codexqa/review-branch-before-merge-with-html-report/ ## How to scan a branch for bugs and security issues before merging with Codexqa Repo: [openqa-cn/codexqa](https://github.com/openqa-cn/codexqa) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README Get a P0 to P3 HTML scan that merges SAST, lint, and secrets results with a semantic pass. **Who it is for:** A developer who just had an agent write code and wants to catch bugs or secrets that a quick glance would miss. **The problem:** AI-written code can hide logic defects, dangerous patterns, or leaked secrets that look correct at first read. **The outcome:** A P0 to P3 HTML defect report with the location, evidence, and a suggestion for each finding. ### Steps 1. Install the defect analyzer with `npx skills add openqa-cn/codexqa --skill codexqa-defect-analyzer`. 2. Check out the branch or paste the newly written code into the agent session. 3. Run the agent prompt below against the branch. 4. Open the HTML scan and sort findings by priority, starting with P0. ### Prompt to paste into your agent ```text Use codexqa-defect-analyzer on this branch in . Merge SAST, lint, and secrets results with one semantic pass and write a P0 to P3 HTML report. Verify that each finding shows its location, evidence, and suggestion. ``` ### Caveat The README names SAST, lint, and secrets as inputs but does not list the specific tools or versions behind them, so check what your environment provides. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/openqa-cn-codexqa/scan-branch-for-bugs-and-security-issues/ ## How to map a diff's impact and generate tests for uncovered paths with Codexqa Repo: [openqa-cn/codexqa](https://github.com/openqa-cn/codexqa) | Difficulty: Intermediate | Time: 45 minutes | Basis: From the README Get an impact report and runnable test files for the paths a change left untested. **Who it is for:** A developer or QA engineer preparing a PR who needs to know what else the change touches and what lacks tests. **The problem:** Changes reach old APIs and call chains that no existing test covers, and reviewers rarely see those paths. **The outcome:** One self-contained HTML change-impact report with a coverage verdict, plus new runnable test files for the uncovered points. ### Steps 1. Install the change analysis skill with `npx skills add openqa-cn/codexqa --skill codexqa-change-analysis`. 2. Check out the branch that contains the diff. 3. Run the agent prompt below. 4. Open the HTML report and review the coverage verdict and sensitive paths. 5. Run the new test files the skill added, and confirm they are counted in the report. ### Prompt to paste into your agent ```text Use codexqa-change-analysis on the diff between and the current branch in this repo. Write one HTML change-impact report and add runnable test files for uncovered points without editing existing test files. Verify that the report lists which methods changed, the entries that reach them, and the uncovered paths. ``` ### Caveat The README says existing test files are not edited, but the new test files still need to be run by you, and the README does not state which test runner they target. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/openqa-cn-codexqa/find-untested-paths-from-one-diff/ ## How to find PRD gaps and conflicts before anyone writes code with Codexqa Repo: [openqa-cn/codexqa](https://github.com/openqa-cn/codexqa) | Difficulty: Beginner | Time: 15 minutes | Basis: From the README Get one P0/P1 register of gaps and conflicts from a PRD, with verification steps. **Who it is for:** A product owner or QA lead who has a PRD or story pack and wants it checked before development starts. **The problem:** Gaps and conflicts in requirements cause failed tests later, and they are cheapest to fix before code exists. **The outcome:** A single assignable P0/P1 gap-and-conflict register, where high-priority items include executable verification steps. ### Steps 1. Install the requirement analyzer with `npx skills add openqa-cn/codexqa --skill codexqa-requirement-analyzer`. 2. Place the PRD or story pack in the repository or paste it into the agent session. 3. Run the agent prompt below. 4. Assign the P0 and P1 rows to the owners named in your team process. ### Prompt to paste into your agent ```text Use codexqa-requirement-analyzer on the PRD at . Produce one P0/P1 gap-and-conflict register, with preconditions, stimulus, expected result, and evidence for each high-priority row. Verify that every high-priority row has an executable verification step. ``` ### Caveat The README describes the register format but does not show a sample PRD, so the quality of gap detection on your own document is unverified until you run it. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/openqa-cn-codexqa/find-requirement-gaps-before-coding/ ## How to replay a browser flow and check the evidence it shows with Codexqa Repo: [openqa-cn/codexqa](https://github.com/openqa-cn/codexqa) | Difficulty: Advanced | Time: 60 minutes | Basis: From the README Run a UI case in Playwright Chromium and get a report with marked screenshots per step. **Who it is for:** A QA engineer who needs to confirm that a web flow reaches the visible result named in its goal. **The problem:** An agent saying a flow is done does not prove the page shows the expected result. **The outcome:** A reports//report.html file showing each step, the control used, and a marked screenshot with the step result. ### Steps 1. Install the browser skill with `npx skills add openqa-cn/codexqa --skill codexqa-jev-browser`. 2. Install Playwright Chromium in the environment, as the README names it as the browser; the README does not give the exact command. (inferred, check the README) 3. Prepare a YAML, Markdown, or API case, or write a goal in plain text. 4. Run the agent prompt below. 5. Open reports//report.html and check each marked screenshot against the expected evidence. ### Prompt to paste into your agent ```text Use codexqa-jev-browser to replay the case at against . Write a report under reports//report.html with a marked screenshot per step. Verify that the final screenshot shows the visible evidence named in the goal. ``` ### Caveat The README says replay does not call a decision model, but generating a case from a goal does, and the README does not say how it behaves on logins or dynamic pages. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/openqa-cn-codexqa/replay-browser-flow-and-check-visible-evidence/