claude-desktop-extra: Skill for Claude Code

.claude/skills/fresh-upstream/SKILL.md

fresh-upstream is a skill for Claude Code from patrickjaja/claude-desktop-extra. It costs 83 tokens per session (1,021 once invoked), scanned C, original, MIT.

A procedure for obtaining and unpacking a clean, unmodified copy of the latest official Claude Desktop Linux package for patch analysis. It verifies the package using signed repository metadata, GPG, and SHA256 checksums.

In plain words
What is it for?
Use it to compare the tracked version with the latest release, download and verify the package, and extract its app.asar into the repository's temporary directory.
Why use it?
It gives developers a trustworthy upstream baseline for comparing patches, instead of comparing against a bundle that was already modified.

Skill for Claude Code

Written for Claude Code: allowed-tools in frontmatter.

This is patrickjaja/claude-desktop-extra's own configuration. It tells Claude Code how to work on claude-desktop-extra itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything claude-desktop-extra configures →

Needs its repository: it runs a file that does not travel with it, so clone the repository first. The line is rm -rf ./tmp/extract ./tmp/app.asar.contents ./extract ./build/work 2>/dev/null || true.

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/fresh-upstream/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.

agentmods badge for fresh-upstream

README.md
[![agentmods](https://agentmods.dev/badge/skills/patrickjaja/claude-desktop-extra/fresh-upstream/github.svg)](https://agentmods.dev/skills/patrickjaja/claude-desktop-extra/fresh-upstream)
Your own site
<a href="https://agentmods.dev/skills/patrickjaja/claude-desktop-extra/fresh-upstream"><img src="https://agentmods.dev/badge/skills/patrickjaja/claude-desktop-extra/fresh-upstream/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.

agentmods 80×15 button for fresh-upstream

Your own site · 80×15
<a href="https://agentmods.dev/skills/patrickjaja/claude-desktop-extra/fresh-upstream"><img src="https://agentmods.dev/badge/skills/patrickjaja/claude-desktop-extra/fresh-upstream.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 83 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,021 The whole file, excluding the scripts and references it only reads on demand.
Security scan C 1 finding. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector warn 7 Sept 2026
SkillSpector: 3 findings, up to high

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 →

  • high Tool Misuse · line 5
    Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).
    Fix: Validate all tool parameters against an allowlist. Reject dangerous parameter values (shell=True, --force, -rf /) and use safe defaults.
  • high Tool Misuse · line 41
    Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).
    Fix: Validate all tool parameters against an allowlist. Reject dangerous parameter values (shell=True, --force, -rf /) and use safe defaults.
  • medium Rogue Agent · line 5
    Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.
    Fix: Remove any persistence mechanisms (cron jobs, startup scripts, state files). Skills should not maintain state across sessions without explicit user consent.
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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00083 $0.01021
Opus 5 $0.00042 $0.00511
Sonnet 5 $0.00017 $0.00204
Haiku 4.5 $0.00008 $0.00102

Measured 9d ago against content hash 427b57ac3b83, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade C, and why

fresh-upstream scanned grade C with 1 finding 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 9d 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.

Recursive force deletehighDestructive command

rm -rf with a variable or a broad path is one typo away from removing the wrong tree.

allowed-tools: Bash(rm -rf *), Bash(mkdir -p *), Bash(ar *), Bash(tar *), Bash(dpkg-deb *), Bash(asar *), Bash(node --check *), Bash(ls *), Bash(cat *), Bash(stat *), Bash(.github/scripts/apt-fetch-verify.sh *), Bash(sha
.claude/skills/fresh-upstream/SKILL.md · 69 lines

How it starts

The opening of the file, as written. The whole thing — 69 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Fresh upstream extract (clean, unpatched)

Goal: a pristine unpatched bundle in ./tmp/ so patch work compares against the true upstream. Run from the repo root. ./tmp/ is gitignored.

Critical: scripts/build-patched-tarball.sh runs apply_patches.py - its output is PATCHED. For a clean baseline use the manual fetch + asar flow below.

Steps

  1. cd to the repo and read tracked version:

    cd "$(git rev-parse --show-toplevel)"
    cat .upstream-version
    
  2. Resolve latest upstream version (GPG + SHA256-verified via the canonical fetcher):

    LATEST_VERSION=$(.github/scripts/apt-fetch-verify.sh poll)
    echo "upstream latest: $LATEST_VERSION"
    

    Report latest vs .upstream-version so the user sees whether this is a new version.

  3. Download + verify the official .deb if missing or stale. apt-fetch-verify.sh download resolves the highest amd64 version, downloads the .deb into ./tmp/, and verifies its SHA256 against the signed Packages index (which is itself verified against the GPG-signed Release):

    mkdir -p ./tmp
    DEB=$(.github/scripts/apt-fetch-verify.sh download amd64 ./tmp)
    echo "verified .deb: $DEB"
    

    (~150MB.) If the download or verification fails, stop and report - do NOT trust an unverified .deb.

  4. Wipe old extracted bundles (project-local only - never touch ~/.config/Claude):

    rm -rf ./tmp/extract ./tmp/app.asar.contents ./extract ./build/work 2>/dev/null || true
    

    Leave the freshly downloaded ./tmp/claude-desktop_*_amd64.deb in place and leave ./build/ packages alone unless the user asked for a full clean.

  5. Extract clean, unpatched to ./tmp/. Crack the .deb (dpkg-deb -x, or ar + tar if dpkg-deb is unavailable), then asar extract the app.asar:

    dpkg-deb -x "$DEB" ./tmp/extract
    asar extract ./tmp/extract/usr/lib/claude-desktop/resources/app.asar ./tmp/app.asar.contents
    
    • Unpatched main bundle: ./tmp/app.asar.contents/.vite/build/index.js
    • Renderer bundles: ./tmp/app.asar.contents/.vite/renderer/*/assets/*.js
    • ion-dist (3P-config SPA): ./tmp/extract/usr/lib/claude-desktop/resources/ion-dist/
    • i18n: ./tmp/extract/usr/lib/claude-desktop/resources/*.json

Read the full file on GitHub · 69 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.

  1. 9d ago First seen · 69 lines · 83 tokens per session scan C 427b57ac3b83

Subscribe to this mod's changes

fresh-upstream is a skill published in the GitHub repository patrickjaja/claude-desktop-extra (622 stars, last pushed yesterday), licensed MIT. It adds 83 tokens to every session and 1,021 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other skills, from other repositories

bastra-recall

Persistent external brain for ChatGPT, Codex, Claude, and other MCP clients — documents (PDFs, contracts, scans with OCR), personal facts (appointments, decisions, items, amounts), AND code lessons / preferences / project topology. USE PROACTIVELY in three modes. (1) RECALL — whenever the user asks about anything from…

n0mad-ai/bastra-recall · 0 tokens

bastra-recall

Proactive private local memory for ChatGPT and Codex — recall before acting, save durable rules, lessons and decisions without being asked. Requires the local bastra-recall MCP server installed by bastra install codex.

n0mad-ai/bastra-recall · 50 tokens

calibre-distill-topic

Synthesize ONE topic across MULTIPLE Calibre books into a single, concept-keyed agent skill — original multi-source authorship (decision frameworks, cross-source config tables, an explicit where-authors-agree/disagree section), with a full ISBN bibliography that doubles as a live-source binding for later deep dives.…

caelum29/calibre-mcp · 160 tokens

calibre-mcp-setup

Install and verify the calibre-mcp server, end to end. Guides an agent from 'Calibre is installed' to a verified working MCP connection: preflight (Node, calibredb, Content Server — macOS/Windows/Linux commands), client-specific install (Claude Code plugin or claude mcp add, Claude Desktop .mcpb or JSON config, Cowork…

caelum29/calibre-mcp · 194 tokens

calibre-study-notes

A tool that creates study notes from books in a Calibre library for a human learner. It uses questions, recall exercises, source pointers, and optional Anki export rather than only providing a summary.

caelum29/calibre-mcp · 243 tokens

himalayas-employer

AI hiring assistant for Himalayas. Post remote jobs, manage company profiles, source talent, message candidates, and benchmark salaries using the Himalayas MCP server.

Himalayas-App/himalayas-mcp · 37 tokens