Skip to content

docs: a fourth rule a first implementation gets wrong - #4

Merged
tactino merged 1 commit into
mainfrom
docs/reconnect-rule
Sep 11, 2026
Merged

tactino merged 1 commit into
mainfrom
docs/reconnect-rule

Conversation

@tactino

@tactino tactino commented Sep 11, 2026

Copy link
Copy Markdown
Member

The protocol tour listed three rules a first implementation usually gets wrong. There is a fourth, and it cost a real run.

A reconnect starts from nothing. Everything the server knows about an environment lives for exactly one connection. A client that reconnects and sends the feedback it was holding puts a transition built from an empty observation into the training buffer, and neither side reports an error.

Added to both the English and Chinese pages, headings renumbered. mkdocs build --strict passes.

Specified in PlugRL/plugrl-protocol#3 as SPEC §7.6; server-side fix in PlugRL/plugrl-server#4.

🤖 Generated with Claude Code

The page listed three rules a first implementation gets wrong. There is a
fourth, and it cost a real run: everything the server knows about an
environment lives for exactly one connection, so a client that reconnects
and sends the feedback it was holding puts a transition built from an empty
observation into the training buffer. Neither side reports an error.

Specified in plugrl-protocol SPEC.md section 7.6; the server-side half is
plugrl-server#4.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
@tactino
tactino merged commit 3cc2c66 into main Sep 11, 2026
1 check passed
@tactino
tactino deleted the docs/reconnect-rule branch September 11, 2026 19:49
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant