DSH voice/SFX alert: plays 'done' on turn end and 'fail' on errors — per-turn dedup, error-first, three modes (cloned-voice TTS / 20 built-in SFX / off), voice library with clone & preset routing + trial synthesis, waveOut/MCI players never touching system volume, credential-free control. 每轮结束自动播报完成/失败：内置 20 音效零配置，也可克隆自己的音色。
Install verification: runtime verified (L5) · tested on dsh 0.2.1-rc.3 · last checked Install verification details →
Data as of (UTC)
What it doesAI
DSH voice/sound alert plugin that auto-announces each turn and plays a failure sound on errors, with 20 built-in sounds and voice cloning.
- Auto voice/sound alert at the end of every turn
- Plays a failure cue when an error occurs
- 20 built-in sounds (10 alerts + 10 nature), zero config; optional voice cloning
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
dsh plugin --profile web add dsh-voice-alertInstall method: npm · runtime verified (L5) · verified on dsh 0.2.1-rc.3
Verification & Compatibility
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
Based on a single engines.dsh range (decides alone; peerDependencies is not consulted). raw ranges
>=0.1.7-rc.1
✓ Running dsh 0.2.1-rc.5 is within the declared range — install is not blocked by version.
Deciding channel: since 0.2.1-alpha.4 DSH judges by engines.dsh when declared — peerDependencies is then not consulted.
Whether it actually runs is a separate question → see the measured matrix below (L1–L5).
| plugin ＼ dsh | dsh 0.2.1-rc.3 | dsh 0.2.1-rc.2 | dsh 0.2.1-alpha.2 | dsh 0.2.0-rc.2 | dsh 0.2.0-rc.1 |
|---|---|---|---|---|---|
| 0.4.18 | · | · | · | · | · |
| 0.4.15 | ✓ | · | · | · | · |
| 0.4.13 | · | ✓ | · | · | · |
| 0.4.12 | · | · | · | · | · |
| 0.4.11 | · | · | · | · | · |
0.4.18npm2026-10-30
dsh-voice-alert
No runtime install test for this version (static scan only) — compatibility unknown, never assumed.
0.4.15npm✓ 1 runtime-verified2026-10-26
dsh-voice-alert
✓dsh 0.2.1-rc.32026-10-26L5 · runtime verified
- • Node.js: v24.14.0
- • DSH: 0.2.1-rc.3
- • Plugin artifact: npm · dsh-voice-alert@0.4.15
ID: dsh-voice-alert-2@0.4.15@dsh0.2.1-rc.3 · profile: l4-sandbox
Issued: 2026-10-26 · expires: 2026-11-02
View checks & evidence
↳ carried from earlier record dsh-voice-alert-2@0.4.18@dsh0.2.1-rc.5 (l3-fullscan)
↳ carried from earlier record dsh-voice-alert-2@0.4.18@dsh0.2.1-rc.5 (l3-fullscan)
↳ carried from earlier record dsh-voice-alert-2@0.4.18@dsh0.2.1-rc.5 (l3-fullscan)
active in loader tree (150 entries)
L1–L3 are plugin-level static conclusions (baseline, identical across dsh versions): evidence is stored once per plugin.
0.4.13npm✓ 1 runtime-verified2026-10-21
dsh-voice-alert
✓dsh 0.2.1-rc.22026-10-21L5 · runtime verified
- • Node.js: v24.14.0
- • DSH: 0.2.1-rc.2
- • Plugin artifact: npm · dsh-voice-alert@0.4.13
ID: dsh-voice-alert-2@0.4.13@dsh0.2.1-rc.2 · profile: l3-fullscan
Issued: 2026-10-21 · expires: 2026-10-28
View checks & evidence
↳ carried from earlier record dsh-voice-alert-2@0.4.18@dsh0.2.1-rc.5 (l3-fullscan)
↳ carried from earlier record dsh-voice-alert-2@0.4.18@dsh0.2.1-rc.5 (l3-fullscan)
↳ carried from earlier record dsh-voice-alert-2@0.4.18@dsh0.2.1-rc.5 (l3-fullscan)
active in loader tree (150 entries)
L1–L3 are plugin-level static conclusions (baseline, identical across dsh versions): evidence is stored once per plugin.
0.4.12npm2026-10-18
dsh-voice-alert
No runtime install test for this version (static scan only) — compatibility unknown, never assumed.
0.4.11npm2026-10-13
dsh-voice-alert
No runtime install test for this version (static scan only) — compatibility unknown, never assumed.
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/dsh-voice-alert-2/?ref=badge)``[](https://www.dsh.so/artifact/dsh-voice-alert-2/?ref=badge-install)``[](https://www.dsh.so/artifact/dsh-voice-alert-2/?ref=badge-install)``[](https://www.dsh.so/artifact/dsh-voice-alert-2/?ref=badge-install)``[](https://www.dsh.so/artifact/dsh-voice-alert-2/?ref=badge-install)``[](https://www.dsh.so/artifact/dsh-voice-alert-2/?ref=badge-install)`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.
v0.4.18dsh manifest0.4.18npm2026-10-300.4.18same versionclean · npmVet is an AST static scan (npm artifact or git source).
show static findings (11)
- warningchildProcessImport · code-exec
lib/edgevoice.js:23import { spawn, spawnSync } from "node:child_process"; - warningchildProcess · code-exec
lib/generate.js:82const probe = spawnSync("ffmpeg", ["-version"], { windowsHide: true, stdio: "ignore" }); - warningchildProcess · code-exec
lib/generate.js:94child = spawn(ffmpegPath, args, { windowsHide: true, stdio: ["ignore", "pipe", "pipe"] }); - warningchildProcess · code-exec
lib/generate.js:206const result = spawnSync( - warningchildProcess · code-exec
lib/index.js:478const probe = spawnSync(config.pythonPath, [String(config.volumeProbeScript)], { - warningnetIp · network
lib/index.js:835const url = new URL(req.url || "/", "http://127.0.0.1"); - warningnetIp · network
lib/index.js:1264const url = new URL(req.url || "/", "http://127.0.0.1"); - warningnetIp · network
lib/index.js:1340const url = new URL(req.url || "/", "http://127.0.0.1"); - warningchildProcess · code-exec
lib/player.js:265const child = spawn(command.file, command.args, LAUNCH_OPTIONS); - warningchildProcess · code-exec
lib/player.js:620const run = spawnSync(command, ["-c", "import sys;print(sys.executable)"], { - warningchildProcess · code-exec
lib/player.js:636const where = spawnSync("where", ["python"], { encoding: "utf8", windowsHide: true, timeout: 6000 });
show 3 audit findings
- mediumLoopback-only POST routes parse any content type, exposing state-changing endpoints to cross-site requests
lib/index.jsThe routes /settings (POST), /generate, /voices/save, /voices/import and /voices/probe authorise solely on req.socket.remoteAddress and never inspect Content-Type or Origin/Sec-Fetch-Site. A malicious web page can therefore send a simple cross-origin POST to http://127.0.0.1:<port>/dsh-voice-alert/settings with content-type text/plain and a JSON body (no CORS preflight) and the handler will parse and apply it: flipping `enabled`/`alertMode`, overwriting the three alert texts, replacing the whole voice library, and forcing real TTS synthesis via /generate or /voices/probe that spends the user's
- lowFirst 8 characters of the TTS API key are returned over HTTP and written to status.json
lib/index.jsstatusPayload()/settingsPayload()/voicesLibraryPayload() expose a 8-character prefix of the long-lived vendor credential; statusPayload is also persisted to disk by control.writeStatus(), so any local reader of status.json (or any process that can reach the loopback routes) learns part of the secret, measurably reducing its effective entropy.
- lowCross-origin GET requests can trigger arbitrary audible playback and monopolise the audio device
lib/index.js/play, /preview and /sfx/play are side-effecting GET endpoints guarded only by the loopback check; any web page the user visits can issue no-cors GETs to repeatedly launch the powershell/python player, spamming audio and (via acquirePlaybackSlot/occupyPlaybackSlot) blocking legitimate cues for seconds at a time. Impact is nuisance/harassment rather than data loss.
Defensive, well-structured plugin; main risk is an unauthenticated loopback HTTP surface whose POST routes parse any content type, allowing cross-site config tampering and forced TTS spending.
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 2026-10-25 · activity: Active
• Repo created: 2026-09-16
• Stars: ★ 1 · Forks: 0
• Health: Active — committed within last 30 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 dsh-voice-alert?
dsh-voice-alert is a DeepSeek Harness (dsh) plugin listed on dsh.so — an independent community registry that indexes public source repositories and publishes automated security-scan and install-verification results for each entry.
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 dsh-voice-alert?
Run dsh plugin --profile web add dsh-voice-alert in a DeepSeek Harness (dsh) environment. That exact command passed a real install inside an isolated sandbox (L4) plus web smoke testing (L5).
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 2026-10-30 (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.