A full review workflow for the claude-desktop-extra project. It compares project patches with the official Linux package, checks documentation and Linux compatibility, examines logs, and combines findings from several review agents.

Use it for a broad project audit or to review a specific area such as patches, compatibility, documentation, or runtime behavior.

It helps reveal outdated patches, missing changes, inaccurate instructions, and problems across different Linux distributions and desktop sessions. Evidence and recommended actions are collected in one report.

How patrickjaja/claude-desktop-extra configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to patrickjaja/claude-desktop-extra. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/patrickjaja/claude-desktop-extra/master/.claude/skills/audit/SKILL.md

Clone the repo

git clone --depth 1 https://github.com/patrickjaja/claude-desktop-extra

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/skills/patrickjaja/claude-desktop-extra/audit/github.svg)](https://agentmods.dev/skills/patrickjaja/claude-desktop-extra/audit)

Your own site

<a href="https://agentmods.dev/skills/patrickjaja/claude-desktop-extra/audit"><img src="https://agentmods.dev/badge/skills/patrickjaja/claude-desktop-extra/audit/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/patrickjaja/claude-desktop-extra/audit"><img src="https://agentmods.dev/badge/skills/patrickjaja/claude-desktop-extra/audit.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 67 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 1,176 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits

NVIDIA SkillSpector pass 7 Sept 2026

SkillSpector: 1 finding, up to low

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

low Excessive Agency · line 11
Skill's behavior or capabilities extend beyond its stated purpose. Scope creep allows an agent to perform actions unrelated to its documented functionality, increasing the attack surface.

Fix: Limit the skill's scope to its documented purpose. Remove instructions that enable the agent to perform actions outside its stated functionality.

How audits are shown

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.00067	$0.01176
Opus 5.5	$0.00027	$0.00470
Sonnet 5.5	$0.00013	$0.00235
Haiku 4.5	$0.00007	$0.00118

Measured 1mo ago against content hash 438e9d75437d, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

audit scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/skills/audit/SKILL.md · 36 lines

How it starts

The opening of the file, as written. The whole thing — 36 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Audit - orchestrated project review

You are the coordinator. Spawn a team of sub-agents to review in parallel, then synthesize ONE report. Run from the repo root. $ARGUMENTS may scope the audit (e.g. "cowork only", "patches only"); default = full.

Use the AskUserQuestion tool (not limited to 4) whenever a finding needs the user's call. Read /architecture and /linux context first if unsure of the domain.

Pre-flight (you do this before fan-out)

Ensure a fresh unpatched upstream extract exists for patch comparison: if ./tmp/app.asar.contents/.vite/build/index.js is missing or older than today, run /fresh-upstream (or tell the user and run the manual .deb crack + asar flow). Patches must be judged against the true current upstream, not a stale extract.
Note .upstream-version vs the highest version in the official apt Packages index. Check ~/.config/Claude/logs/ exists for the log workstream.

Fan out - spawn these workstreams as parallel agents (Explore/general-purpose)

Give each agent the project path, the fresh-extract path, and ask for findings + file:line evidence + a recommended action, not file dumps.

Patches vs upstream. For each patches/*/*.nim: does its target still exist in the fresh ./tmp bundle? Run ./scripts/validate-patches.sh ./tmp/app.asar.contents and report pass/fail. For failures, classify: pattern renamed / code refactored / feature removed / feature upstreamed (→ needs regression guard). Flag patches that may be vestigial (target gone, no longer needed) and gaps (new darwin/win32 gates with no Linux patch - diff process.platform old vs new). Verify EXPECTED_PATCHES strictness + no false-success idempotency (AGENTS.md Rule 6).
Docs accuracy. Cross-check baseline/CLAUDE_FEATURE_FLAGS.md, CLAUDE_BUILT_IN_MCP.md, ION.md, PLATFORM_GATE_BASELINE.md, README.md patch table, and AGENTS.md against the actual current bundle + patches. Flag stale minified names, wrong counts, removed features still documented. (Read CHANGELOG.md head only: offset 1, limit 60.)
Linux compatibility. Walk the support matrix (X11, Wayland-wlroots, Wayland-GNOME, Wayland-KDE, XWayland) × (Arch/Ubuntu/Debian/Fedora/RHEL/NixOS/Jetson, x86_64+aarch64). Check input (js/cu_linux_executor.js, executor_linux.js) + screenshot cascades still cover each session, the glibc floor (2.34 overall; kwin-portal-bridge 2.39) holds, launcher session detection is sound. Surface edge cases (ydotoold version, Niri, immutable distros, sandboxed/portal identity).
Cowork backend (bundled native VM). Cowork runs on the .deb's bundled native VM backend (cowork-linux-helper + virtiofsd + smol-bin + QEMU/OVMF; requires /dev/kvm). The old claude-cowork-service Go daemon is deprecated and out of scope. Verify the Cowork-related patches/*/*.nim still match their targets in the fresh ./tmp bundle and that any regression guards still assert the upstreamed behavior. Flag drift against the current bundle.
Runtime logs (if present). Scan ~/.config/Claude/logs/{main,cowork_vm_node,mcp,claude.ai-web}.log for errors/exceptions, permission denials, dispatch bridge issues (DISPATCH-FWD/DISPATCH-TRANSFORM), and renderer crashes. cowork_vm_node.log is the bundled backend's Electron-side log (still valid). Defer deep dispatch/cowork debugging to /debug.

Read the full file on GitHub · 36 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

1mo ago Changed 438e9d75437d
2mo ago First seen · 36 lines · 67 tokens per session scan A 715e96ff182c

Subscribe to this mod's changes

audit is a skill published in the GitHub repository patrickjaja/claude-desktop-extra (726 stars, last pushed yesterday), licensed MIT. It adds 67 tokens to every session and 1,176 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Code review Debugging anthropic appimage arch-linux aur claude claude-desktop

audit next to 6 skills from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
auditpatrickjaja/claude-desktop-extra	Skill	726	yesterday	MIT	A	67	Claude Code
oraclesteipete/oracle	Skill	4.1k	yesterday	MIT	A	18	Claude Code, Codex
semgrep-rule-creatorwaybarrios/opencode-power-pack	Skill	551	4d ago	MIT	A	37	Claude Code
variant-analysiswaybarrios/opencode-power-pack	Skill	551	4d ago	MIT	A	45	Claude Code
code-review-webrampstackco/claude-skills	Skill	1.0k	11d ago	MIT	A	121	Claude Code
ln-44-performance-optimizerlevnikolaevich/claude-code-skills	Skill	587	yesterday	MIT	A	21	Claude Code
ln-71-operations-investigatorlevnikolaevich/claude-code-skills	Skill	587	yesterday	MIT	A	24	Claude Code

claude-desktop-extra: Skill for Claude Code