Skip to content

fix: LMM session affinity and native Pi file tool guidance - #12

Merged
LIghtJUNction merged 4 commits into
mainfrom
fix/cache-routing
Oct 1, 2026
Merged

LIghtJUNction merged 4 commits into
mainfrom
fix/cache-routing

Conversation

@LIghtJUNction

@LIghtJUNction LIghtJUNction commented Oct 1, 2026 •

Copy link
Copy Markdown
Contributor

LMM Chat Completions models such as GPT-Pro / gpt-6-astra inherited vendor compat without the gateway session-affinity flag. The gateway now enables affinity for all admitted Chat Completions models; explicit provider/model false overrides still reach the real adapter. Long-retention defaults stay scoped to reviewed routes.

LMM models can also assume Codex apply_patch exists in bash, leaving an unwritten artifact that a later preview tries to open. A before_agent_start hook now describes only Pi tools that are actually active, directs file writes/edits through their supplied schemas, and asks the model to recover and verify a file before previewing after failure. For local built-in bash, a tool_call guard checks direct apply_patch commands in the configured shell and blocks them when unavailable. A standalone literal xdg-open to a missing local file is blocked before execution. Existing apply_patch commands and custom/remote tool environments are preserved. Older Pi hosts without an effective shell-settings API retain prompt guidance; current Pi 0.99.1 was probed through its real native RPC loader and exposes the API and builtin bash metadata. Prompt guidance remains for compound preview commands and custom shell prefixes; this is not a general shell parser. It installs no shell command.

Validation: all 102 tests pass, including real Pi adapter session headers/default/false overrides and native write/edit creation/update/failure behavior and actual ExtensionRunner tool_call dispatch through native bash, including no downstream command after block and no preview after failed write; typecheck and real Pi 0.99.1 native-loader smoke pass. No credentials, models.json, parent submodule pin, or production configuration changed.

Copilot AI balanced review requested due to automatic review settings October 1, 2026 02:33
@chatgpt-codex-connector

Copy link
Copy Markdown

You have reached your Codex usage limits for security reviews. Please try again later.

Copilot AI left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.

@LIghtJUNction
LIghtJUNction merged commit 2692cb0 into main Oct 1, 2026
3 checks passed
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.

2 participants