Skills pr-walkthrough

📦

pr-walkthrough

Name: pr-walkthrough
Author: warpdotdev

Content revision r1 Medium Risk ⚡ Contains scripts⚙️ External commands🌐 Network access📁 Filesystem access

Explore Pull Requests with Interactive D3 Maps

Complex pull requests are difficult to understand from diffs alone. This skill creates a validated visual walkthrough of architecture, data flow, dependencies, and user actions.

Supports: Claude Codex Code(CC)

📊 67 Adequate

Install with my Agent

Copy this request to your Agent. It includes the canonical Skill page and manifest.

Agent request

Review the Skillstore skill "pr-walkthrough" from https://skillstore.io/skills/warpdotdev-pr-walkthrough.md and its manifest at https://skillstore.io/api/skills/warpdotdev-pr-walkthrough/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.

Your Agent should still show its plan and request any confirmation required by the security policy.

Agent-readable resources

Use these links when an AI agent, crawler, or script needs clean context instead of reading the full page.

Markdown detail GET /skills/warpdotdev-pr-walkthrough.md Signed manifest GET /api/skills/warpdotdev-pr-walkthrough/manifest Signed lockfile GET /api/skills/warpdotdev-pr-walkthrough/lockfile

Test it

Using "pr-walkthrough". Create a compact walkthrough for the pull request on this branch.

Expected outcome:

A local interactive walkthrough with four switchable views, guided tours, linked evidence, and a completed browser validation result.

Using "pr-walkthrough". Help reviewers understand how this feature moves data through the system.

Expected outcome:

A concise subsystem overview for architectural orientation.
A directed data-flow map linked to relevant pull request evidence.
A guided sequence explaining state ownership and rendered outcomes.

Using "pr-walkthrough". Prepare a walkthrough for a large cross-layer pull request.

Expected outcome:

A validated visual guide covering architecture, dependencies, user actions, specifications, review comments, and important areas for reviewer inspection.

Security Audit

Medium Risk

v1 • 9/4/2026 Open versioned report

Most static alerts are false positives caused by JavaScript template literals, Markdown formatting, keyboard events, CSS, and normal D3 rendering. Confirmed risks are unsanitized link schemes in dynamic detail HTML and the third-party D3 runtime dependency. PR descriptions and comments also require stronger prompt-injection boundaries.

Files scanned

1,081

Lines analyzed

Review items

False positives ignored

Confirmed security concerns (1)

Medium

SKILL.md:69-75

Untrusted PR Content Enters the Agent Context

The workflow imports pull request bodies and human or agent comments as source material. It limits code-change instructions but does not reject other embedded override or tool-use requests.

The collection and use of external comments is explicit, and line 75 provides only a partial trust-boundary rule. Prompt injection remains possible through other instructions in that content.

Capability review items (3)

These are real local capabilities that may be expected for this skill, so they require review but are not counted as confirmed malicious behavior.

Medium

scripts/d3_canvas_runtime.py:162

innerHTML assignment (XSS risk)

panel.innerHTML = `

Dynamic text is escaped, but URL values are inserted into href attributes without a scheme allowlist. An attacker-influenced javascript: URL could execute when a reviewer clicks it.

Low

scripts/d3_canvas_runtime.py:13

Hardcoded URL

D3_CDN_URL = f"https://cdn.jsdelivr.net/npm/d3@{D3_VERSION}/dist/d3.min.js"

Generated walkthroughs load executable D3 code from jsDelivr at runtime. The version is pinned, but the remote script lacks a documented integrity check.

Low

SKILL.md:19

Hardcoded URL

- `https://cdn.jsdelivr.net/npm/d3@7.9.0/dist/d3.min.js`

The instructions require generated pages to load D3 from a third-party CDN. This creates an intentional network and remote-code supply-chain dependency.

Risk Factors

⚡ Contains scripts (2)

scripts/d3_canvas_runtime.py:155 scripts/d3_canvas_runtime.py:162

⚙️ External commands (50)

🌐 Network access (10)

scripts/d3_canvas_runtime.py:13 scripts/validate_d3_canvas.py:70 scripts/validate_d3_canvas.py:71 SKILL.md:17 SKILL.md:121 SKILL.md:226 SKILL.md:19 SKILL.md:127 SKILL.md:269 SKILL.md:273

📁 Filesystem access (1)

scripts/d3_canvas_runtime.py:289

Audited by: codex

Share & cite this report

Share the versioned assessment report, neutral badge, embed card, and citations. Skillstore reports evidence without deciding whether this Skill is safe.

Open versioned report

Copy report link

https://skillstore.io/skills/warpdotdev-pr-walkthrough/audits/1?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Markdown badge

[![Skillstore security assessment](https://skillstore.io/badges/skills/warpdotdev-pr-walkthrough/security.svg)](https://skillstore.io/skills/warpdotdev-pr-walkthrough?utm_source=security_passport_badge)

HTML badge

<a href="https://skillstore.io/skills/warpdotdev-pr-walkthrough?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/warpdotdev-pr-walkthrough/security.svg" alt="Skillstore security assessment" loading="lazy"></a>

Embed card

<iframe src="https://skillstore.io/embed/skills/warpdotdev-pr-walkthrough.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>

Academic citations (APA · BibTeX · CFF)

APA citation

warpdotdev. (2026). pr-walkthrough security audit report (audit version 1) [Author version unspecified]. Skillstore. https://skillstore.io/skills/warpdotdev-pr-walkthrough/audits/1

BibTeX citation

@techreport{warpdotdev-warpdotdev-pr-walkthrough-2026,
  author = {warpdotdev},
  title = {pr-walkthrough security audit report (audit version 1)},
  institution = {Skillstore},
  year = {2026},
  number = {1},
  url = {https://skillstore.io/skills/warpdotdev-pr-walkthrough/audits/1},
  note = {Author version unspecified}
}

CITATION.cff

cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "pr-walkthrough security audit report (audit version 1)"
version: "unspecified"
type: report
authors:
  - name: "warpdotdev"
date-released: "2026-09-04"
url: "https://skillstore.io/skills/warpdotdev-pr-walkthrough/audits/1"
identifiers:
  - type: other
    value: "skillstore:warpdotdev-pr-walkthrough:audit:1"
    description: "Skillstore immutable audit report identifier"

Skillstore Score

Why this score Evidence Confidence: Medium

Architecture

Maintainability

Content

Community

Spec Compliance

What You Can Build

Orient a pull request reviewer

Create a compact visual guide before reviewing a change across several related files.

Explain subsystem impact

Show how a proposed change connects architecture, data movement, dependencies, and visible behavior.

Prepare an implementation walkthrough

Generate a navigable artifact for discussing a pull request with distributed collaborators.

Try These Prompts

Create a basic walkthrough

Create an interactive walkthrough for the pull request on my current branch. Keep it compact and validate the generated canvas.

Focus on reviewer orientation

Build a PR walkthrough for my current branch. Emphasize the stable subsystem architecture and the smallest set of concepts needed for review.

Trace behavior across the change

Generate the four PR walkthrough views. Prioritize data movement, dependency direction, user actions, changed specifications, and existing review discussion.

Create and publish a validated walkthrough

Create a detailed walkthrough for pull request [number]. Validate every view, then publish through approved protected hosting without exposing private repository content.

Best Practices

Use the full current files and adjacent modules to establish architectural context.
Keep each view focused on distinct reviewer questions and the smallest useful node set.
Validate every graph, tour, control, link, and browser-rendered state before sharing.

Avoid

Do not treat the walkthrough as a substitute for code review findings.
Do not build architecture explanations from diff hunks alone.
Do not publish private pull request content without explicit approval and suitable access controls.

Frequently Asked Questions

What does this skill create?

It creates a static interactive HTML walkthrough with four D3 views and guided reviewer tours.

Does it review the code?

No. It explains the change and surrounding architecture without generating new approval recommendations or review findings.

Can the walkthrough run without a development server?

Yes. The output opens from a local file URL, although D3 normally loads from a pinned CDN.

What repository information does it use?

It uses pull request metadata, diffs, current source files, changed specifications, tests, review comments, and relevant visual assets.

How is the result validated?

The included validator checks required graphs and controls, then uses Playwright to verify browser rendering and interactions.

Can I publish the walkthrough?

Yes, after explicit approval. Use protected hosting for private code and confirm that the artifact contains no sensitive information.

Developer Details

Author

warpdotdev

License

MIT

Skillstore revision

Version notice

The author did not declare a version.

Repository

https://github.com/warpdotdev/common-skills/tree/b811c24365ae505bfc9646458957b886e29110b5/.agents/skills/pr-walkthrough

Ref

e5464e662405de6361fdde6b984f9ea865635f64

Maintenance freshness

9/8/2026

Usage

0 downloads · 0 views

File structure

📁 scripts/

📄 d3_canvas_runtime.py

📄 validate_d3_canvas.py

📄 SKILL.md