Repository navigation
perf(setup): download an org in about 2s instead of 30s - #93
Merged
Merged
Conversation
The wizard's All download took 31.6s on an org of about 300 resources; almost all of it was fixed waiting, not network. - Skip pull's 2s --force pause for setup's own pulls (VAPI_GITOPS_FORCE_PAUSE=off). The wizard ran one --force pull per type. - Stop applying the 700ms request gap to GETs; 429s are still retried with backoff. Writes keep the gap. - When All is picked, run one full pull instead of one per type. - Create resources/<org>/ before the wizard's first pull, as non-interactive setup does, so the org pin is still written. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
scott-lowe-vapi
marked this pull request as ready for review
November 7, 2026 00:07
dhruva-vapi
approved these changes
Nov 7, 2026
Contributor
Author
Merge activity
|
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Value
V.A.L.U.E. tier: micro. It's a mechanical change of 36 lines across 3 files, and the before/after timings below prove it.
npm run setup, including everynpx create-vapi-appuser, plus anyone runningnpm run pull.How it works
--forcepause for setup's own pulls.pull --forcewarns and waits 2s so someone typing it by hand can Ctrl+C. The wizard ran one--forcepull per resource type, so 9 types meant 18s of pauses. setup now setsVAPI_GITOPS_FORCE_PAUSE=offfor the pulls it starts, following theVAPI_GITOPS_ORG_PIN=offconvention. A hand-typed--forcestill pauses.api.tswaited 700ms between every request, reads included, which was about 9 of a plain pull's 11 seconds. Reads now skip the gap. A 429 is still retried with backoff (5 tries from 2s), as the config-freevapi-client.tsalready does. Writes (POST, PATCH, DELETE) keep the gap.--force --skip-bindingspull. "Let me pick…" still pulls by--idper type..vapi-org.json) is only written into a folder that already exists. Non-interactive setup createsresources/<org>/first, but the wizard didn't, so the pin used to come from its second per-type pull. With a single pull, nothing would have written it, so the wizard now creates the folder first too.Verification
These are the exact command sequences each path runs, timed against a real org (
scott-dev: 8 tools, 13 assistants, 8 structured outputs, 2 squads, 11 personalities, 118 scenarios, 118 simulations, 11 suites, 8 evals). They used read-only pulls in a throwaway copy of the repo.npm run pullnpm run pull -- <org> --force.vapi-org.jsonincluded.npm run buildpasses, andnpm testpasses 584/584.🤖 Generated with Claude Code