dsh-llm-aws-kiro
Run the following command in DeepSeek Harness:
dsh plugin install joshryandavis/dsh-llm-aws-kiro
Paste the following prompt into your AI chat to install this plugin:
Install it in your DeepSeek Harness by running dsh plugin install joshryandavis/dsh-llm-aws-kiro; the full source is available at https://github.com/joshryandavis/dsh-llm-aws-kiro .
About this plugin
If you already signed in through kiro-cli or the Kiro IDE but want to call the 15 models behind Kiro straight from a DeepSeek Harness, this plugin bridges the gap. As a standalone Cordis plugin it mounts into any harness composition via cordis.yml, registers the kiro provider route on ctx.llm, and requires zero changes to the harness itself. It picks up credentials ambiently from kiro-cli's SQLite store or the Kiro IDE SSO cache, silently refreshes expired tokens over AWS SSO OIDC or the Kiro desktop auth service, and writes the result back so both consumers stay in sync; a CI or bot deployment can skip ambient discovery entirely with a bearerTokenEnv reference. On the wire it speaks the AWS event-stream protocol with Smithy framing, enforces capacity backoff, first-token and idle watchdogs, and maps the harness off/low/high/max reasoning vocabulary onto each model's authenticated effort ladder while attaching Kiro thinking markers to the system prompt. A shared management-catalog cache keeps model capacities and context windows in step with the per-credential, per-region discovery data. It is built for developers who already hold a Kiro login and want Claude 4.7/4.8, DeepSeek V3.2, MiniMax M2.5, GLM-5, Qwen3 Coder, and friends flowing through a DeepSeek Harness, as well as for automation pipelines that prefer a static bearer token.
Use Cases
- Call Claude 4.8, DeepSeek V3.2, GLM-5 and 12 more Kiro-backed models directly inside a DeepSeek Harness workflow
- Drive Kiro inference from CI/CD pipelines using a static bearer token with no interactive login
- Plug an existing kiro-cli or Kiro IDE session into the Harness without a second sign-in
- Mix and match providers in one Harness while sharing a unified retry, timeout, and capacity-backoff policy
Best For
- Developers who already have a kiro-cli or Kiro IDE login and want to reuse it inside the Harness
- Platform and toolchain engineers wiring Kiro inference into CI/CD without interactive auth
- Advanced users extending DeepSeek Harness with custom Cordis plugin bundles
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.