by V1ki
Use ChatGPT (Codex), Claude, and Grok (X Premium) subscriptions as DeepSeek Harness LLM providers — OAuth login in the web UI, no API keys
# Add to your Claude Code skills
git clone https://github.com/V1ki/dsh-plugin-subscriptionsGuides for using ai agents skills like dsh-plugin-subscriptions.
Last scanned: 8/19/2026
{
"issues": [],
"status": "PASSED",
"scannedAt": "2026-08-19T04:36:02.283Z",
"npmAuditRan": false,
"pipAuditRan": true,
"promptInjectionRan": true
}See how dsh-plugin-subscriptions compares with popular alternatives.
dsh-plugin-subscriptions is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by V1ki. Use ChatGPT (Codex), Claude, and Grok (X Premium) subscriptions as DeepSeek Harness LLM providers — OAuth login in the web UI, no API keys. It has 616 GitHub stars.
Yes. dsh-plugin-subscriptions passed SkillsLLM's automated security scan — a dependency vulnerability audit plus prompt-injection heuristics — with no high-severity issues. You can read the full report in the Security Report section on this page.
Clone the repository with "git clone https://github.com/V1ki/dsh-plugin-subscriptions" and add it to your Claude Code skills directory (see the Installation section above).
dsh-plugin-subscriptions is primarily written in TypeScript. It is open-source under V1ki on GitHub, so you can review or fork the full source.
Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh dsh-plugin-subscriptions against similar tools.
No comments yet. Be the first to share your thoughts!
From $49 · 30 days · terms & refunds
⚠️ Third-Party Software Notice
This skill is third-party open-source software developed and hosted independently on GitHub. SkillsLLM is an informational directory and does not control or maintain the underlying repository.
Any security checks, ratings, or warnings displayed by SkillsLLM are automated and limited in scope. They do not constitute a security certification or guarantee that the software is safe, error-free, or free from malicious code, vulnerabilities, compromised dependencies, or prompt-injection risks.
Review the source code, permissions, dependencies, and configuration before installing or running any third-party skill. Use is at your own risk. To the maximum extent permitted by applicable law, SkillsLLM is not liable for losses arising from third-party software.
Automated security scan — not a certification.
See comparison
English | 中文
Use your ChatGPT (Codex), Claude, Grok (X Premium), GitHub Copilot, and Google Antigravity subscriptions as LLM providers in DeepSeek Harness — no API keys. Codex, Grok, and Antigravity log in via OAuth in the dsh web UI (Settings → Subscriptions), while Copilot uses the GitHub OAuth device flow; Claude imports credentials from an existing Claude Code session when there is one (macOS Keychain or ~/.claude/.credentials.json) and otherwise falls back to the same browser OAuth flow, so the Claude Code CLI is not required. Tokens live at ~/.dsh/plugins/subscriptions/auth.json (mode 0600) and refresh automatically.
Open Settings → Subscriptions → Manage to set a subscription display name, or edit a model's display name under Edit model list. Both share the existing Save/Cancel actions; clearing a name or choosing Reset to default and saving restores the original name. Names are single-line text of up to 80 characters, stored per provider and model ID in provider-settings.json, and survive restarts and catalog refreshes. They only change presentation: provider IDs, model IDs, reasoning settings, and account routing remain unchanged. Existing account aliases continue to work independently.
Subscription display names also label provider groups in the native model picker. Saving a cleared name restores the provider's default group title.
Antigravity now advertises one picker entry per effort group, using the same native model/effort controls as Codex. When supplied, agentModelSorts alone determines the selectable roster/order; only explicit Low/Medium/High variants present there are grouped. tieredModelIds category keys such as pro and flashLite are not model identities and do not expand the picker. Catalog thinkingLevel and thinkingBudget settings are preserved on the wire. Explicit Low/Medium/High suffixes in known model families provide a compatibility fallback for older catalogs; unrelated models are not merged by display name.
Group defaults prefer High, then Medium, then Low when no valid saved preference exists. Existing concrete model IDs remain callable. Account pools retain their existing conservative intersection of members' efforts. Refresh the model list after installing so older caches acquire roster metadata. Without roster metadata, legacy/internal filtering is best effort; exact parity with the official client requires checking the live account. See verification notes.
Antigravity usage keeps explicitly reported shared limits from retrieveUserQuotaSummary. The Gemini group includes Argon; native two-group summaries report weekly and five-hour windows for Gemini and Claude/GPT. Account-wide shared quotas remain visible independently of model picker visibility. Settings and the composer dialog show remaining percentages with horizontal bars, without duplicating effort variants. Missing shared groups can supplement uncovered model quotas from fetchAvailableModels; a reset timestamp alone never establishes a cadence. Supplemental failures retain valid summaries; cancellation propagates.
Settings → Subscriptions: per-provider login/logout, no API keys. Claude imports credentials from Claude Code when available and otherwise uses OAuth, as Codex and Grok always do (settings screenshots use demo accounts and catalog data):

Configure visibility, default reasoning effort, and context together in Edit model list, with shared Save and Cancel actions:

Configure image generation, video generation, and X search per provider. Tool switches apply only to sessions created after saving:

Logged-in providers join the session model picker with their live model catalogs:

Models that advertise reasoning levels get an Effort selector in the same menu — Codex models, Grok 4.6 / 4.5, and Copilot's reasoning models (levels and defaults come from each provider's live catalog, not a hardcoded list; Copilot's capabilities.supports.reasoning_effort array is sent as reasoning_effort on chat completions and reasoning.effort on the Responses wire). Models listing both Copilot endpoints (gpt-5.4, gpt-5-mini) normally speak chat completions but reroute to /responses when a request combines function tools with an effort — Copilot rejects that combination on the chat wire:

Codex models whose catalog advertises the fast tier (the codex CLI's fast mode) get a Speed toggle in the composer's tool row, next to the model selector — Standard or Fast (service_tier: priority), per session. The /fast slash command offers the same choice as a popup; it errors with an explanation when the current model has no fast tier.
Subagents have a separate per-model default: Settings → Subscriptions → ChatGPT (Codex) → Manage → Edit model list → Subagent default speed. Save changes to apply it to subsequent requests, including running subagents. This setting persists across restarts and works regardless of the parent model; main-session Speed choices do not propagate to children. A speed explicitly selected in a child session overrides its model default, including Standard. Explicit session choices reset on restart. Fast still requires catalog support for the actual Codex model. Hosts without live agent origin metadata keep Standard unless the session has an explicit speed choice.

The composer's stats row gains a subscription usage pill showing quota information for the provider of the session's current model (Codex for a Codex model, Grok for a Grok model, and so on). It shows at most one provider: when switching to a non-subscription model, it keeps the most recent subscription selected in the mounted conversation view, or stays hidden if there is none. This recent-model history is not persisted across page reloads. Click the pill to expand every logged-in provider and account — the default account is starred, and the current provider is listed first. Antigravity previews only the current model's windows (at most two per account); the other model windows remain available in a closed disclosure.
Settings → Subscriptions → Status-bar quota display selects Current / most recent subscription only (default) or Hidden. The display preference is saved in the current browser, applies immediately, survives reloads, and synchronizes between tabs of the same origin. It does not hide usage details in Settings, change model selection, or change account routing.
Quota percentages independently selects Used (default) or Remaining for Settings, the composer/status-bar pill, and its expanded account rows. Both preferences are browser-local and synchronize across tabs. Bars fill in the selected direction, while warning colors still indicate high used quota (80% warning, 95% critical). Unknown or invalid telemetry stays unknown, never 100% remaining; passed reset times retain the last reported value until a refresh.
For multiple accounts, the compact pill says e.g. “Codex · 2 signed-in accounts”; expand it for each account's windows, model scopes and reset times. Percentages are never added or averaged. Signed-in accounts need not be members of the selected model's pool. A single-account readout labels the default/listed account; neither a default star nor session affinity proves which account successfully served a request.
Expanded details and Settings show the last successful fetch time, stale status and the next permitted retry after a failed refresh, when known. Unsupported quota queries, empty results, and failed queries without a previous snapshot have distinct messages. If an older backend omits quota timestamps, the UI shows when it received the response and explicitly says the quota update time was not provided; receipt time never proves that upstream quota was refreshed. Clicking the account Refresh button requests force: true, but endpoint failure cooldowns still apply. Opening the badge rereads normally, not forcibly. Existing snapshots survive failures without being relabeled fresh; the endpoint cooldown remains intact. Open Settings and badges share successful usage reads in connection-local memory and reread that memory on focus, without additional upstream requests. This does not poll the server cache on focus; normal badge polling and manual refresh retain their existing behavior. The preference never changes internal usedPercent, account selection, exhaustion protection or cooldowns. Pool membership and successful-serving-account telemetry remain separate follow-ups to #202. Antigravity summary endpoints/grouping in #182, #184 and #187 are not incorporated here.

The image_generate tool renders its r