llm-relay 0.3.0 → 0.3.1
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +3 -0
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -291,6 +291,9 @@ subagent happens to read redirect its own routing. Both cases are covered by tes
|
|
|
291
291
|
Precedence for a subagent request: `@relay:` directive → `subagents[<tier>]` →
|
|
292
292
|
`subagents.default` → normal routing. Omit `routing.subagents` entirely and nothing changes.
|
|
293
293
|
|
|
294
|
+
📄 Full design, the wire evidence behind it, and **how to re-verify the marker after a Claude Code
|
|
295
|
+
upgrade**: [docs/subagent-routing.md](docs/subagent-routing.md).
|
|
296
|
+
|
|
294
297
|
Each provider is `kind:"openai"` (translated Anthropic↔OpenAI via llm-bridge) or
|
|
295
298
|
`kind:"anthropic"` (forwarded as-is). In `repair` mode an openai target reshapes on itself;
|
|
296
299
|
an anthropic provider has no fixed model id to reshape on, so it needs an explicit top-level
|
package/package.json
CHANGED