Conversation
* feat(transport): replace shared chat transport with mothership-stream module * improvement(contracts): regenerate contracts from go * feat(tools): add tool catalog codegen from go tool contracts * feat(tools): add tool-executor dispatch framework for sim side tool routing * feat(orchestrator): rewrite tool dispatch with catalog-driven executor and simplified resume loop * feat(orchestrator): checkpoint resume flow * refactor(copilot): consolidate orchestrator into request/ layer * refactor(mothership): reorganize lib/copilot into structured subdirectories * refactor(mothership): canonical transcript layer, dead code cleanup, type consolidation * refactor(mothership): rebase onto latest staging * refactor(mothership): rename request continue to lifecycle * feat(trace): add initial version of request traces * improvement(stream): batch stream from redis * fix(resume): fix the resume checkpoint * fix(resume): fix resume client tool * fix(subagents): subagent resume should join on existing subagent text block * improvement(reconnect): harden reconnect logic * fix(superagent): fix superagent integration tools * improvement(stream): improve stream perf * Rebase with origin dev * fix(tests): fix failing test * fix(build): fix type errors * fix(build): fix build errors * fix(build): fix type errors * feat(mothership): add cli execution * fix(mothership): fix function execute tests
Co-authored-by: Theodore Li <theo@sim.ai>
This was referenced Apr 14, 2026
Merged
6 tasks
waleedlatif1
added a commit
that referenced
this pull request
Jul 14, 2026
…pilot tool executions (#5679) * fix(tools): resolve {{ENV_VAR}} references in user-only params for copilot tool executions Chat-invoked integration tools with API-key auth have been broken since the mothership tool-dispatch rewrite (#4090) removed the orchestrator's env reference resolution. Agents pass {{VAR}} references (the VFS exposes env var names only), and the literal placeholder was sent to the provider, producing auth failures like Sentry's 401 Invalid token. Resolution is deliberately narrower than the removed deep resolver: only whole-value references on params declared visibility user-only, gated to copilot executions, resolving against the same personal+workspace env merge workflow runs use. LLM-writable params (urls, headers, bodies) never resolve, so references cannot be used to extract secrets. Missing variables fail fast with an actionable error instead of a provider-side 401. * refactor(tools): delegate copilot env reference resolution to the shared executor resolver Replaces the hand-rolled exact-match regex with resolveEnvVarReferences (allowEmbedded: false), the same resolver used by workflow runs, MCP config, and webhooks — one set of reference semantics instead of two that can drift. Behavior is identical; all existing tests pass unchanged. * fix(tools): fail fast on env references without user context, clarify personal-only scope errors Addresses review: a copilot execution missing userId now errors explicitly instead of forwarding the literal placeholder upstream, and a missing-variable error without a workspace context explains that only personal variables are in scope there (matching workflow-run resolution semantics).
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
4 participants
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.
Summary
Overhaul of mothership code system with improved typing. New features like docx editing, file streaming, removal of respond structures, etc. Cleanup of BullMQ worker code.
Type of Change
Testing
Tested on staging environment.
Checklist