dsh-plugin-slashx-gateway
Run the following command in DeepSeek Harness:
dsh plugin install superslash-rico/dsh-plugin-slashx-gateway
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install superslash-rico/dsh-plugin-slashx-gateway to install the SlashX gateway plugin into your current Harness instance; the source is at https://github.com/superslash-rico/dsh-plugin-slashx-gateway
About this plugin
DeepSeek Harness is a local-first AI agent runtime, but its native API does not speak the SlashX protocol out of the box. If you want SlashX clients, mobile apps, or third-party integrations to call the models and toolchains loaded inside Harness through a single standardized slashx.request.v1 / slashx.response.v1 interface, without hand-rolling auth, media persistence, SSRF guards, and rich-component fallback yourself, this gateway plugin exists for exactly that gap.
It sits between Harness and SlashX as a secure, idempotent transport bridge: text flows as native content blocks; images are downloaded, decoded, validated, and injected into the model; audio, video, PDFs, Office files, and arbitrary documents are safely written to a per-conversation workspace where agent tools can read them. On the response side, the slashx_deliver tool lets the agent deliver images, video, audio, files, citations, cards, and buttons in protocol-compliant form; when the client does not declare a capability, content is explicitly downgraded to an attachment or Markdown rather than silently dropped. Long-running tasks complete via a one-time callback token, and billing supports both flat-per-run and token pass-through meters with a fail-closed guarantee when usage data is incomplete.
It is best suited for developers and teams who have already built agent applications on DeepSeek Harness and need to expose a standardized inference endpoint to the SlashX ecosystem or to internal and external consumers who expect rich-media interactions. If your threat model calls for SSRF protection, path-traversal guards, token hygiene in logs, and strict capability negotiation, this plugin provides the full gateway layer. The project is still in Developer Preview; run the full acceptance suite in a pre-production environment before any production rollout.
Use Cases
- Serve SlashX chat clients, mobile apps, or third-party integrations through a single standardized inference endpoint
- Let agent toolchains generate and securely deliver images, video, audio, files, and cards
- Settle token-level billing across multi-model runs with a fail-closed guarantee when usage is incomplete
Best For
- Developers building on DeepSeek Harness who need SlashX ecosystem compatibility
- Product teams exposing rich-media inference endpoints with SSRF and path-traversal concerns
- Platform engineering teams with compliance requirements for token metering, async callbacks, and idempotency
Related Plugins
ModLens is a vision plugin for DeepSeek Harness that gives text-only models sight by reading images pasted directly into chat, with zero-config setup and multiple vision engines.
On-demand vision for text-only DeepSeek Harness agents: built-in free keyless vision chain and 14 vision tools, routing image turns as tool calls to vision models with pixel fidelity, no Python needed, one-command install.
Give text-only models in DeepSeek Harness eyes, enabling image Q&A, long-screenshot OCR, UI restoration, and GUI visual tasks.
An armor-breaking plugin for DeepSeek: rewrites instruction priority, expands refusal-blocked phrases, persists across sessions, and shows a green active indicator.