Install verification: Ecosystem app: not installed via dsh plugin add, so the L1-L5 plugin install-verification ladder does not apply; the security scan still runs.Install verification details →
Data as of (UTC)
What it doesAI
DeepSeek Harness native desktop client built on Tauri 2 with bundled Node.js runtime, one-click install, fast startup, tray resident, and auto-updates.
- Built on Tauri 2 for lightweight native desktop app
- Bundled Node.js runtime for seamless execution
- Tray resident and automatic updates
Capability tags are AI-inferred or rule-extracted from the README (see badge) — vocabulary-limited, not install-verified. AI = inferred; README = literal keyword from the readme.
Installation
Ecosystem apps are not installable via dsh plugin add: desktop / web shells install or deploy through their own release channel (GitHub Releases or the project site).
The level reads L5 · Ecosystem plugin / L5 · Ecosystem app per the three-way classification (the L1-L5 ladder does not rate them); the security scan still runs.
Trends
Four lines, four questions: total stars, stars gained this week, cumulative downloads, downloads this week. The GitHub pair comes from real-read sampling — a rotating daily shard, so one plugin is re-read roughly every 5–14 days, sparser than npm; the npm pair has a value on every snapshot day. The two cumulative panels are smoothed curves with no sample markers; the two weekly panels keep their raw bars and add a 4-week moving-average trend line that strips single-week noise down to the direction of travel.
10 samples (roughly every 7 days), 5 of them reconstructed (dashed runs); a smoothed curve through every sample — not a daily series
09-17 → 11-13 · 10 points
Bars = the delta between two samples 5–14 days apart (negative points down, faint bars contain reconstructed points); line = 4-week moving-average trend
09-24 → 11-13 · 9 points
Reconstructed backwards from the current total using daily downloads (gaps are not interpolated)
Only 0 points (< 3): no chart yet
no data yet
Bars = the snapshot’s last-week rolling window (official npm + npmmirror combined); line = 4-week moving-average trend — the shape of weekly demand, not a user count
Only 0 points (< 3): no chart yet
no data yet
Verification & Compatibility◼ Ecosystem app
Grouped by plugin version: each card lists the dsh versions actually tested against that artifact version and their install verdicts (L5 wins). Untested combinations are omitted — absence means untested, never assumed compatible. L1–L3 are a plugin-level static archive (independent of the dsh version — the same verdict on every dsh row): they sit in each row's level grid next to L4/L5, marked "static", while their evidence is stored once per plugin inside that row's "checks & evidence" instead of being duplicated per version.
External credentials (plugin-level): none detected
| plugin ＼ dsh | dsh 0.1.5-rc.2 | dsh 0.1.3-alpha.2 | dsh 0.1.3-alpha.1 | dsh 0.1.2-rc.1 | dsh 0.1.2-alpha.1 |
|---|---|---|---|---|---|
| v1.5.5 | ◼ | ◼ | ◼ | · | · |
v1.5.5github◼ 3 ecosystem app2026-09-13
https://github.com/xiincs/deepseek-harness-desktop
◼dsh 0.1.5-rc.22026-09-13Ecosystem app
- • Node.js: v24.14.0
- • DSH: 0.1.5-rc.2
- • Plugin artifact: github · https://github.com/xiincs/deepseek-harness-desktop#v1.5.5
ID: deepseek-harness-desktop-17@v1.5.5@dsh0.1.5-rc.2 · profile: l4-sandbox
Issued: 2026-09-13 · expires: 2026-09-20
View checks & evidence
not present in web-profile loader inventory
L1–L3 are plugin-level static conclusions (baseline, identical across dsh versions): evidence is stored once per plugin.
◼dsh 0.1.3-alpha.22026-09-09Ecosystem app
- • Node.js: v24.14.0
- • DSH: 0.1.3-alpha.2
- • Plugin artifact: github · https://github.com/xiincs/deepseek-harness-desktop#v1.5.5
ID: deepseek-harness-desktop-17@v1.5.5@dsh0.1.3-alpha.2 · profile: l4-sandbox
Issued: 2026-09-09 · expires: 2026-09-16
View checks & evidence
not present in web-profile loader inventory
L1–L3 are plugin-level static conclusions (baseline, identical across dsh versions): evidence is stored once per plugin.
◼dsh 0.1.3-alpha.12026-09-07Ecosystem app
- • Node.js: v24.14.0
- • DSH: 0.1.3-alpha.1
- • Plugin artifact: github · https://github.com/xiincs/deepseek-harness-desktop#v1.5.5
ID: deepseek-harness-desktop-17@v1.5.5@dsh0.1.3-alpha.1 · profile: l4-sandbox
Issued: 2026-09-07 · expires: 2026-09-14
View checks & evidence
not present in web-profile loader inventory
L1–L3 are plugin-level static conclusions (baseline, identical across dsh versions): evidence is stored once per plugin.
Only the latest 3 published versions are shown; source-snapshot (commit hash) groups, unstated groups and older versions are omitted — the full archive stays in the data layer (snapshot/unstated groups appear only when a plugin has no published version).
Author badges
Embed the dsh.so badges in your README to show this plugin's automated security-scan and install-test status:
`[](https://www.dsh.so/artifact/deepseek-harness-desktop-17/?ref=badge)`dsh.so is an independent community project, not affiliated with or endorsed by DeepSeek AI. Badges are maintained by dsh.so; embed links carry a ?ref= attribution parameter.
Security Report
Two automated layers, neither a manual review: a static scan (per-plugin rule checks) and a model-driven dynamic security audit that covers only the subset static-flagged at warn or above and never executes plugin code.
Static low; dynamic medium (authoritative when present) → overall medium.
v1.5.9v1.5.9git2026-11-14v1.5.9same versionclean · gitVet is an AST static scan (npm artifact or git source).
show static findings (10)
- warningcurlDownload · network
scripts/fetch-node.mjs:65`$ProgressPreference='SilentlyContinue'; Invoke-WebRequest -Uri '${archiveUrl}' -OutFile '${archivePath}'`, - warningnetIp · network
src-tauri/src/lib.rs:4//! Security model: the harness page (http://127.0.0.1:<port>) is a plain - warningnetIp · network
src-tauri/src/lib.rs:483.unwrap_or_else(|| format!("http://127.0.0.1:{}", server::default_port())); - warningnetIp · network
src-tauri/src/lib.rs:1045.unwrap_or_else(|| format!("http://127.0.0.1:{}", server::default_port())); - warningchildProcess · code-exec
src-tauri/src/server.rs:544let mut output = cmd.stdout(Stdio::piped()).stderr(Stdio::piped()).spawn().map_err(|e| { - warningchildProcess · code-exec
src-tauri/src/server.rs:562thread::spawn(move || { - warningchildProcess · code-exec
src-tauri/src/server.rs:568thread::spawn(move || { - warningnetIp · network
src-tauri/src/server.rs:10//! printed `dsh web: http://127.0.0.1:<port>/?token=<secret>` line, keeping - warningnetIp · network
src-tauri/src/server.rs:75const URL_PREFIX: &str = "dsh web: http://127.0.0.1:"; - warningnetIp · network
src-tauri/src/server.rs:864/// dsh web: http://127.0.0.1:3080/?token=<secret> (0.1.5-rc.1)
show 6 audit findings
- mediumExternal-link guard uses substring match, so crafted URLs load inside the harness webview
src-tauri/src/lib.rsNavigationStarting/NewWindowRequested only cancel navigations that pass this test, so a URL such as `https://evil.example/?x=localhost` or `http://127.0.0.1.attacker.tld/` is treated as internal and the harness webview navigates to a fully attacker-controlled document. That document receives the always-injected file-mention script and the WebView2 host-message channel, letting it emit forged `open-file-mention` events into the shell, and it strands the UI away from the harness (back/reload were deliberately removed from the context menu).
- lowFile-mention host bridge authenticates with a constant and forwards an unvalidated remote path
src-tauri/src/lib.rs`source: "dsh-desktop"` is a constant embedded in the script that runs in the same untrusted webview, and any script in that webview can call `window.chrome.webview.postMessage` directly without a click, so the check is not an authentication boundary. The path then arrives as a Tauri event consumed by the shell's IPC (get_editable_preview / open-with / reveal paths). Because the harness server already runs with the user's filesystem rights this grants no new capability on its own, but it makes the remote page the source of a path that shell-side code must sanitise.
- lowPer-boot dsh auth token is written to stdout and the persistent desktop log
src-tauri/src/server.rsThe ready line parsed by extract_ready_url carries `?token=<secret>`, and the stdout reader pushes every line through push_log into the in-memory ring and the persistent `desktop.log` (also returned to the UI by get_log_tail); set_running additionally prints the tokenised URL. The bearer token for the local harness server — which gates the whole UI and /api/* — is therefore stored in a log file readable by any same-user process and exposed to the shell page instead of being kept only in memory.
- lowPlugin install runs arbitrary package specs named by an unvetted crowd-submitted catalog
src-tauri/src/server.rsinstall_plugin forwards the `package` string verbatim to `dsh plugin add`, which resolves it through pnpm — the code's own comments state the catalog it comes from (awesome-dsh-plugin.com) is crowd-submitted and unvetted, and the spec form also accepts `github:owner/repo#path`. A malicious or typosquatted entry therefore executes third-party install scripts as the user; the only gate is the confirm dialog showing the raw string.
- lowBundled Node.js archive is downloaded and executed without integrity verification
scripts/fetch-node.mjs:109The build script fetches the Node archive over HTTPS, extracts it into src-tauri/resources/runtime (a path that ships inside the packaged app) and immediately executes the extracted binary, with no check against nodejs.org's SHASUMS256.txt or any pinned hash. A poisoned mirror/network path or a substituted archive is executed and then bundled, so the integrity of the shipped runtime rests entirely on TLS.
- lowVersion argument is interpolated into a PowerShell -Command string
scripts/fetch-node.mjs:65`version` (process.argv[2]) is embedded into both the URL/path and the single-quoted PowerShell script text, so a value containing a quote (or `$(...)`) breaks out of the literal and runs arbitrary PowerShell at build time; the same interpolation appears in the Expand-Archive call. Reachability is limited to whoever invokes the script (developer/CI, or a wrapper such as the not-harvested prepare-runtime.mjs), hence low.
Tauri shell is well fenced; residual risks: substring external-URL check lets attacker pages into the bridged harness webview, auth token written to logs, and unverified build/plugin downloads.
Rule (2026-10-28): when a dynamic conclusion exists it is authoritative (it may raise or lower the level); the static scan tier (vet primary, dsh backstop) applies only when there is none (not queued / audit failed / withdrawn). The audit never writes the exemption list (exemptions are manual only); a wrong conclusion is withdrawn by human review and the tier falls back to the static scan tier.
Disclaimer: the static scan and the dynamic security audit are both automated analysis, not a security guarantee. Always review what you install.
Activity
Last commit 2026-09-21 · activity: Stable
• Repo created: 2026-08-13
• Stars: ★ 63 · Forks: 5
• Health: Stable — committed within 90 days
Source
Author README
View on GitHub ↗The author's original README from the repository — fetched live, unedited (the English page loads the default README).
Frequently asked questions
What is deepseek-harness-desktop?
deepseek-harness-desktop is an ecosystem app listed on dsh.so, not a DSH plugin package: it cannot be installed with dsh plugin add and ships through the project's own release channel.
Is dsh.so or this plugin page official?
dsh.so is an independent community registry and is not affiliated with, endorsed by, or sponsored by DeepSeek AI.
How do I install deepseek-harness-desktop?
This is an ecosystem app, not a plugin installable via dsh plugin add: desktop / web shells install or deploy through their own release channel. The security scan still runs, but the L1-L5 plugin install ladder does not rate it.
How does the security scan relate to a manual audit?
Both layers are automated, not a manual audit. The static scan is heuristic analysis of the repository source, so it can produce false positives and miss context. The model-driven dynamic security audit only covers plugins static-flagged at warn or above, never executes plugin code, and its conclusion is authoritative when there is one (it may raise or lower the static tier); exemptions stay manual. Neither layer is an endorsement: review the repository and the permissions it asks for before installing, together with the verification level and health signals.
How fresh is the data on this page?
The data on this page is as of 2026-11-14 (UTC) — the most recent valid timestamp among the registry refresh, install verification, security scan and dynamic audit. Build time is never used as a data date.