DevAgentRadar → Goose

You are helping me adopt a real coding-assistant change. Work only from the facts below. Do not invent features.

Context

Assistant: Goose Proposal: Register TrustedRouter and OpenCode Zen as Goose declarative providers Summary: Goose v1.48.0 adds declarative providers for TrustedRouter, OpenCode Zen, Gondola, SayGM, Lynkr, and PleumRouter, and drives cost tracking from custom provider cost fields. Register those providers, fill cost fields, replace Z.ai GLM-5.2 with GLM-5.3, and drop Gemini OAuth. Primary source: https://github.com/aaif-goose/goose/releases/tag/v1.48.0

Why it matters

v1.48.0’s headline is first-class declarative providers: TrustedRouter, OpenCode Zen gateway, Gondola, SayGM, Lynkr, and PleumRouter. Teams still pointing those gateways at a generic OpenAI-compatible custom endpoint should switch to the declarative entries so Goose can sort configured providers to the top of the list and apply provider-specific behavior. Custom provider cost fields now drive cost tracking, and Azure Foundry can estimate cost via inferred catalog pricing, so empty cost fields under-report spend. Model routing also moved: Z.ai GLM-5.2 is replaced by GLM-5.3, GPT-5.6 sol/terra/luna can route via AWS Bedrock with Codex and Responses API follow-up support, and the Gemini OAuth provider is deprecated. After changing providers, run /new for a fresh session without restarting and confirm the model from recently used models in the chat footer picker. Thirty-five further changelog items were omitted from the release body; do not assume extra provider flags beyond what is listed.

Suggested actions

1. Upgrade Goose to v1.48.0. On Linux ARM64 install the new desktop release packages from https://github.com/aaif-goose/goose/releases/tag/v1.48.0 rather than an older x86 build.
2. Open Goose provider config and add TrustedRouter as a declarative provider instead of a generic OpenAI-compatible endpoint. Then add OpenCode Zen as a declarative gateway provider.
3. For each of Gondola, SayGM, Lynkr, and PleumRouter that we use, replace the custom OpenAI-compatible entry with the new declarative provider so configured providers sort to the top of the provider list.
4. Edit every custom provider definition and fill the cost fields Goose now uses for cost tracking. Change any Z.ai model from GLM-5.2 to GLM-5.3. Remove the deprecated Gemini OAuth provider from our config.
5. In the Goose chat input, run /new to start a fresh session without restarting, then select the new provider/model from recently used models in the chat footer model picker.

Config surfaces this release may change

• MCP servers (high confidence) — check your repo before applying
• background and headless runs — check your repo before applying
• context and compaction settings — check your repo before applying
• hooks — check your repo before applying
• permission rules — check your repo before applying
• skills — check your repo before applying

After you finish

Do not report this as applied to DevAgentRadar. You cannot write the visitor's loop.

Tell the human: open https://devagentradar.com/proposals/goose-v1-48-0-register-trustedrouter-and-opencode-zen-as-goose-declarati and mark Applied, Skipped, or Failed. Proposal id: 7bfe8e98-475f-425f-b90c-4c67a2d08cbf

Your job

1. Restate the change in one sentence.
2. Propose a minimal plan for my repo (or a throwaway pilot).
3. Implement only what I approve; prefer small diffs and tests.
4. Call out risks (permissions, breaking APIs, cost).

Start by confirming you understood the proposal.