---
title: "Review Walkthrough: review, setup & limitations · Undominated.ai"
canonical: https://undominated.ai/skills/trailofbits-review-walkthrough/
description: "Turns the committed diff of the current branch into a standalone HTML page of ordered review steps, each with an explanation and severity-tagged findings anchored to diff lines."
---

# Review Walkthrough: review, setup & limitations · Undominated.ai

> Turns the committed diff of the current branch into a standalone HTML page of ordered review steps, each with an explanation and severity-tagged findings anchored to diff lines.

[← Explore all skills](/skills/)

CODE REVIEW / Trail of Bits

# Review Walkthrough

Turns the committed diff of the current branch into a standalone HTML page of ordered review steps, each with an explanation and severity-tagged findings anchored to diff lines.

 [See setup guidance ↓](#setup)[Original source ↗](https://github.com/trailofbits/skills/tree/8e54050ae63c2f62224c35b1616efe71865b4777/plugins/review-walkthrough/skills/review-walkthrough)

SOURCE REVIEW

 Reviewed 2026-10-22
 Evidence 9 linked sources
 Publisher Trail of Bits
 Licence [CC-BY-SA-4.0 ↗](https://raw.githubusercontent.com/trailofbits/skills/8e54050ae63c2f62224c35b1616efe71865b4777/LICENSE)
 Revision 8e54050ae63c
 [Read what was—and wasn’t—checked ↓](#review)

“Generates an interactive HTML walkthrough for reviewing code changes. Use only when explicitly called.”

 [Trail of Bits · upstream description ↗](https://raw.githubusercontent.com/trailofbits/skills/8e54050ae63c2f62224c35b1616efe71865b4777/plugins/review-walkthrough/skills/review-walkthrough/SKILL.md) Our analysis follows below.

01 / THE REASONING

## Why this made the selection.

 - Its bundled renderer compares every step with the complete captured patch, so a step whose per-file diff was split, rewritten, omitted or invented is refused rather than shown.
- The skill keeps the checkout untouched: it says not to change branches, reset, edit source files, run the project's tests or install its dependencies in order to build the walkthrough.
- It ships in trailofbits/skills; the plugin README names Facundo Tuesca of Trail of Bits as author.

### A good fit for

 - Walk a reviewer through a branch's final diff in concept-sized steps, with definitions before the code that uses them and tests beside their subjects
- Pin high, medium and low findings to a file and line of the displayed diff so they can become inline PR comments
- Prepare a gh api review command for a matching GitHub PR without posting anything

### Weigh up before choosing

 - Runs only on explicit invocation (disable-model-invocation: true): /review-walkthrough:review-walkthrough in Claude Code, $review-walkthrough in Codex.
- Covers committed changes up to the captured HEAD; uncommitted work is left out and only mentioned.
- Inline-comment export needs PR metadata whose head and base match the captured diff; otherwise pr_meta is null and findings stay on the page.
- The page never submits a review. The user runs the copied gh api command; Comment and Request Changes need a summary before it can be copied.
- Findings are the model's reading of the changed code and its surroundings; the skill tells it not to run tests to build the page.
- The renderer needs uv and Python 3.11 or later; gh is needed only for PR metadata and for running the copied command.
- `/plugin marketplace add trailofbits/skills` adds the whole Trail of Bits marketplace; this skill comes with the review-walkthrough plugin chosen afterwards in /plugin menu.
- Last commit touching plugins/review-walkthrough/skills/review-walkthrough at the pinned revision: 2026-09-14.
- Not executed: it was not installed or run.

02 / THE REVIEW RECORD

## What we actually inspected.

Source review has boundaries. A clear record is more useful than a “safe” badge.

### Material inspected

 - plugins/review-walkthrough/skills/review-walkthrough/SKILL.md
- plugins/review-walkthrough/skills/review-walkthrough/scripts/render_walkthrough.py
- plugins/review-walkthrough/skills/review-walkthrough/scripts/pyproject.toml
- plugins/review-walkthrough/skills/review-walkthrough/template.html (searched for scripts, links and URLs)
- plugins/review-walkthrough/skills/review-walkthrough/agents/openai.yaml
- plugins/review-walkthrough/README.md
- plugins/review-walkthrough/.claude-plugin/plugin.json
- README.md (installation and licence sections)
- LICENSE

### Our findings

 - The base is the user's, else the open PR's, else the remote default branch; the skill says not to assume main and to ask when the base is missing or ambiguous.
- The patch is captured with git diff --no-ext-diff --no-textconv between the merge base and HEAD, saved to a temporary file outside the checkout.
- render_walkthrough.py has no third-party dependencies; it reads --input and --diff, refuses an --output equal to either or to the template, and writes one HTML file.
- The renderer runs git apply --numstat -z from the system temp directory to parse patch paths; the code comment says this does not apply the patch.
- template.html has one inline <script> and no external script, stylesheet or fetch; its only URL strings are an XHTML namespace check and a SAFE_HREF pattern that lets fragment links point at http(s), mailto, # or / targets.
- agents/openai.yaml sets allow_implicit_invocation: false, matching the explicit-only frontmatter.

### Not established by this review

 - Not executed: it was not installed or run.
- The renderer, the page and its gh api export were not run.

The review applies to the material and revision named here. A newer upstream release can change its behavior.

03 / PUT IT TO WORK

## Add a skill to your workflow.

[Upstream setup instructions ↗](https://raw.githubusercontent.com/trailofbits/skills/8e54050ae63c2f62224c35b1616efe71865b4777/README.md)
 - In Claude Code, add the marketplace with `/plugin marketplace add trailofbits/skills` (repository README). This adds the whole Trail of Bits marketplace, not review-walkthrough alone.
- Open `/plugin menu` and install the review-walkthrough plugin, which holds this one skill.
- Invoke it explicitly with `/review-walkthrough:review-walkthrough`, optionally naming a base such as "compare this branch against origin/develop" (plugin README).

### Before you start

 - Git
- `uv` and Python 3.11 or later for the bundled renderer
- GitHub CLI (`gh`), optional, for PR metadata and for running a copied review command

### Compatibility

Claude Code · Codex

### Reads the branch diff; writes the review data and one HTML file

 - Run git to resolve the base and merge base and to capture the patch; no branch change, reset or source edit
- Read changed files and the surrounding source, tests, configuration and dependency metadata
- Write the walkthrough JSON, the captured patch and the HTML page outside the checkout unless the user names a location
- Run the bundled Python renderer through uv
- Optional: query GitHub PR metadata through gh; the page only builds a gh api command for the user to copy
- Open the page with open or xdg-open when a browser is available

### Cost model

CC-BY-SA-4.0. The source names no paid service; GitHub access through gh is optional and uses the user's own account.

04 / FOLLOW THE EVIDENCE

## The source trail.

Our notes are separate from the original resource. Check upstream before adopting a new version.

 - [Review Walkthrough SKILL.md ↗](https://raw.githubusercontent.com/trailofbits/skills/8e54050ae63c2f62224c35b1616efe71865b4777/plugins/review-walkthrough/skills/review-walkthrough/SKILL.md) Checked 2026-10-22 https://raw.githubusercontent.com/trailofbits/skills/8e54050ae63c2f62224c35b1616efe71865b4777/plugins/review-walkthrough/skills/review-walkthrough/SKILL.md Supports: upstreamDescription, summary, whySelected, bestFor, limitations, compatibility, access, review.findings
- [Bundled renderer render_walkthrough.py ↗](https://raw.githubusercontent.com/trailofbits/skills/8e54050ae63c2f62224c35b1616efe71865b4777/plugins/review-walkthrough/skills/review-walkthrough/scripts/render_walkthrough.py) Checked 2026-10-22 https://raw.githubusercontent.com/trailofbits/skills/8e54050ae63c2f62224c35b1616efe71865b4777/plugins/review-walkthrough/skills/review-walkthrough/scripts/render_walkthrough.py Supports: access, review.findings
- [Renderer pyproject.toml ↗](https://raw.githubusercontent.com/trailofbits/skills/8e54050ae63c2f62224c35b1616efe71865b4777/plugins/review-walkthrough/skills/review-walkthrough/scripts/pyproject.toml) Checked 2026-10-22 https://raw.githubusercontent.com/trailofbits/skills/8e54050ae63c2f62224c35b1616efe71865b4777/plugins/review-walkthrough/skills/review-walkthrough/scripts/pyproject.toml Supports: install.prerequisites
- [Walkthrough page template ↗](https://raw.githubusercontent.com/trailofbits/skills/8e54050ae63c2f62224c35b1616efe71865b4777/plugins/review-walkthrough/skills/review-walkthrough/template.html) Checked 2026-10-22 https://raw.githubusercontent.com/trailofbits/skills/8e54050ae63c2f62224c35b1616efe71865b4777/plugins/review-walkthrough/skills/review-walkthrough/template.html Supports: review.findings
- [Codex interface metadata agents/openai.yaml ↗](https://raw.githubusercontent.com/trailofbits/skills/8e54050ae63c2f62224c35b1616efe71865b4777/plugins/review-walkthrough/skills/review-walkthrough/agents/openai.yaml) Checked 2026-10-22 https://raw.githubusercontent.com/trailofbits/skills/8e54050ae63c2f62224c35b1616efe71865b4777/plugins/review-walkthrough/skills/review-walkthrough/agents/openai.yaml Supports: review.findings
- [review-walkthrough plugin README ↗](https://raw.githubusercontent.com/trailofbits/skills/8e54050ae63c2f62224c35b1616efe71865b4777/plugins/review-walkthrough/README.md) Checked 2026-10-22 https://raw.githubusercontent.com/trailofbits/skills/8e54050ae63c2f62224c35b1616efe71865b4777/plugins/review-walkthrough/README.md Supports: compatibility, install.steps, install.prerequisites, limitations, access
- [review-walkthrough plugin.json ↗](https://raw.githubusercontent.com/trailofbits/skills/8e54050ae63c2f62224c35b1616efe71865b4777/plugins/review-walkthrough/.claude-plugin/plugin.json) Checked 2026-10-22 https://raw.githubusercontent.com/trailofbits/skills/8e54050ae63c2f62224c35b1616efe71865b4777/plugins/review-walkthrough/.claude-plugin/plugin.json Supports: whySelected
- [trailofbits/skills README ↗](https://raw.githubusercontent.com/trailofbits/skills/8e54050ae63c2f62224c35b1616efe71865b4777/README.md) Checked 2026-10-22 https://raw.githubusercontent.com/trailofbits/skills/8e54050ae63c2f62224c35b1616efe71865b4777/README.md Supports: install, license, compatibility
- [trailofbits/skills LICENSE ↗](https://raw.githubusercontent.com/trailofbits/skills/8e54050ae63c2f62224c35b1616efe71865b4777/LICENSE) Checked 2026-10-22 https://raw.githubusercontent.com/trailofbits/skills/8e54050ae63c2f62224c35b1616efe71865b4777/LICENSE Supports: license

KEEP COMPARING

## Other approaches to consider.

Related by category or shared topics. These are alternatives to inspect, not a measured quality order.

 [### Second Opinion ↗ Captures one patch (uncommitted work, a branch against a base, or a single commit) and has the Codex, Antigravity or Gemini CLI review it, then reports each reviewer's findings and where two reviews agree or differ.](/skills/trailofbits-second-opinion/)[### Django performance review ↗ Review Django code and report only performance issues the reviewer traced. Zero findings is an allowed result.](/skills/getsentry-django-perf-review/)[### Sentry code review ↗ Reviews a change for runtime, query, security, design, and test gaps, and does not block on style.](/skills/getsentry-code-review/)

 [AI Tools ↗](/tools/)[Skills ↗](/skills/)[Agents ↗](/agents/)[MCP Servers ↗](/mcp-servers/)[Workflows ↗](/workflows/)

## Continue your investigation

 - [Choose a toolkit](/tools/)
- [Inspect execution hosts](/agents/)
- [Check tool access](/mcp-servers/)
- [Check project dependencies](/check-repo/)
