Install verification: runtime verified (L5) · tested on dsh 0.2.2-rc.1 · last checked Install verification details →
Data as of (UTC)
What it doesAI
Two AI account pools for DeepSeek Harness: TeleAgent with daily auto check-in and AutoClaw with local-credential discovery and quota-aware failover.
- Dual AI account pools: TeleAgent (TeleAI) and AutoClaw (Zhipu desktop)
- TeleAgent pool performs daily automatic check-in
- AutoClaw discovers local credentials and fails over based on quota
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 github:836157055/dsh-teleagent-poolInstall method: GitHub · runtime verified (L5)
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
package manifest not collected — the declared range is unknown.
| plugin ＼ dsh | dsh 0.2.2-rc.1 current |
|---|---|
| 98ccd0 | · |
| 29a71b | ✓ |
98ccd0github2026-11-10
https://github.com/836157055/dsh-teleagent-pool
No runtime install test for this version (static scan only) — compatibility unknown, never assumed.
29a71bgithub✓ 1 runtime-verified2026-11-10
https://github.com/836157055/dsh-teleagent-pool
✓dsh 0.2.2-rc.1current2026-11-10L5 · runtime verified
- • Node.js: v24.14.0
- • DSH: 0.2.2-rc.1
- • Plugin artifact: github · https://github.com/836157055/dsh-teleagent-pool#29a71bdd9e3802cb44233b28c46db7cd811819b0
ID: dsh-teleagent-pool@29a71bdd9e3802cb44233b28c46db7cd811819b0@dsh0.2.2-rc.1 · profile: l4-sandbox
Issued: 2026-11-10 · expires: 2026-11-17
View checks & evidence
↳ carried from earlier record dsh-teleagent-pool@98ccd0@dsh0.2.2-rc.1 (l3-fullscan)
↳ carried from earlier record dsh-teleagent-pool@98ccd0@dsh0.2.2-rc.1 (l3-fullscan)
↳ carried from earlier record dsh-teleagent-pool@98ccd0@dsh0.2.2-rc.1 (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.
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-teleagent-pool/?ref=badge)``[](https://www.dsh.so/artifact/dsh-teleagent-pool/?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 high; dynamic medium (authoritative when present) → overall medium.
v1.1.0dsh manifest1.1.0git2026-11-111.1.0same versionsuspicious · gitVet is an AST static scan (npm artifact or git source).
show static findings (8)
- criticalsecretSk · secrets
src/product.js:44export const INFER_API_KEY = 'sk-qvp4m6h9g20uM6PgYP7E' - warningnetIp · network
src/autoclaw/bridge.js:44export const DISCOVERY_URL = `http://127.0.0.1:${DISCOVERY_PORT}/` - warningnetIp · network
src/autoclaw/bridge.js:151const response = await fetch(`http://127.0.0.1:${port}/`, { - warningnetIp · network
src/autoclaw/instances.js:48const response = await fetch(`http://127.0.0.1:${port}/`, { - warningshellExec · shell
src/autoclaw/switch.js:203execFile('powershell.exe', [ - warningfsWrite · files
src/autoclaw/switch.js:183fs.writeFileSync(tmp, `${JSON.stringify({ schemaVersion: 1, accountKey })}\n`, { encoding: 'utf8', mode: 0o600 }) - warningfsWrite · files
src/autoclaw/vaultkeeper.js:76fs.writeFileSync( - warningfsWrite · files
src/autoclaw/vaultkeeper.js:109fs.chmodSync(tmp, 0o600)
show 3 audit findings
- mediumPlugin terminates and relaunches the AutoClaw desktop app via a fixed PowerShell command
src/autoclaw/switch.js:248execFile with an argument array and a constant script means there is no shell string interpolation and no attacker-controlled input, so this is not command injection. It is still a privileged capability: the plugin force-kills and restarts the user's running AutoClaw application, which can lose unsaved work and is a process-control action a provider plugin would not otherwise need. Reachability is inferred from the module's stated auto-switch purpose since index.js was not harvested.
- lowCopies AutoClaw's encrypted credential file into the plugin's own data directory, persisting it
src/autoclaw/vaultkeeper.js:72The plugin duplicates another application's authentication material out of AutoClaw's userData into a backup directory and keeps it after the client itself deletes the file. The copy is still the encrypted v10/DPAPI-bound blob (not a plaintext token), so this is not credential disclosure, but it creates a second location of secret material with less-defined lifetime. accountKey is regex-validated (^[a-f0-9]{64}$) so there is no path traversal.
- lowHardcoded vendor client identity keys (shared app key) in source
src/autoclaw/product.js:46The AutoClaw client appKey is embedded in the plugin and used to compute X-Auth-Sign. It is the vendor's client-identity key extracted from their desktop binary, not a user secret, so actual exposure is limited, but it is shared credential material now redistributed inside a third-party plugin. src/product.js also carries a bundled inference channel key (value redacted in the harvested copy).
No injection, backdoor, or third-party exfiltration found; residual concerns are a fixed-argument PowerShell process restart and duplication of AutoClaw's encrypted credentials.
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 2026-11-03 · activity: Active
• Repo created: 2026-11-03
• Stars: ★ 0 · 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 @836157055/dsh-teleagent-pool?
@836157055/dsh-teleagent-pool 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 @836157055/dsh-teleagent-pool?
Run dsh plugin --profile web add github:836157055/dsh-teleagent-pool 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 2026-11-11 (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.