Repository navigation
feat(delivery): support native builds and refresh no-project skills - #88
Merged
polymaniadeveloper-hub merged 7 commits intoNov 5, 2026
Merged
Conversation
Thread transfer impact✅ Thread transfer remains within every enforced ceiling.
Baseline: Scenario and decoded snapshot size10 historical turns, 5 command tools per turn, 878.9 KiB retained MCP result per historical turn, and a 1.05 MiB retained result in the measured turn.
Updated in place by a trusted workflow. PR artifacts are strictly validated and never executed. |
polymaniadeveloper-hub
force-pushed
the
feat/native-team-execution-current
branch
from
November 5, 2026 18:28
abe998f to
9368b05
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No-project composers skipped skill refresh, and the separate team engine could not use native T3 execution. Global catalogs now refresh on web, desktop and mobile. An opt-in native build coordinator delegates coding work through T3 while the engine retains its review and approval gates.
Native results and cancellation are reconciled by the server; durable identities and recovery guards prevent duplicate writers. The first slice requires a coordinator in the engine worktree and does not migrate read-only research or automate attachment. Built-in Kimi, Hermes and DeepSeek settings are registered with managed policy. Tool descriptions are compacted so total metadata stays below the base budget. A scanner regression test now waits for healthy reads before advancing the blocked read's test clock.
Validation: policy/context/native tests 18 passed, server typecheck and touched-file lint passed. Context metadata measures 47,477 bytes versus 47,538 on the base. Windows scanner run: 56 passed, 7 platform skips, 2 existing symlink-permission failures; the changed FIFO case is Linux-only and awaits CI. Earlier native/delivery and provider-skill suites passed. Model-facing eval still hits the baseline initialization error before model calls. The local aggregate checker was interrupted; full CI and live-provider/UI verification remain outstanding.
Created with GPT-6.1 Sol in the Codex harness, including a GPT-6.1 Sol implementation subagent.