problem · Revision 1
## Observed symptom Intermittent HTTP 400 "invalid request error" on /responses - retries do not always recover (deepseek/deepseek-v4.1-flash) I'm on DSH 0.1.7-rc.2. My plan is: GOAT. Username: zjwzeroba12. ## Environment GitHub CommandCodeAI/command-code #970 ## Trigger / contex
problem · Revision 1
## Observed symptom OpenAI Responses: default reasoning effort "none" breaks gpt-6.1-sol (HTTP 400) Selecting `gpt-6.1-sol` on the OpenAI Responses services without configuring `reasoning` makes pipecat send `reasoning={"effort": "none"}`. OpenAI rejects that with HTTP 400, so ev
problem · Revision 1
## Observed symptom flyctl throws a bad error if token refresh fails <!-- prilog-copy-issue:5626910133 --> Copied for testing from https://github.com/superfly/flyctl/issues/5249 ## Literal error text ``` Error: get app: verify: invalid token: all tokens missing third-party discha
problem · Revision 1
## Observed symptom [Bug] Managed compact handoff strips max_tokens → anthropic_messages upstream 400 'body.max_tokens Field required', whole fallback chain fails Managed compact handoff requests sent over `anthropic_messages` fail with a deterministic `400` because `sanitizeComp
problem · Revision 1
## Observed symptom split: an invalid ANTHROPIC_API_KEY ends in a 60-line traceback instead of a one-line error `pr-split split feat/publish-cadence --base main --dry-run --stack` pr-split v0.1.1, default `llm` backend, with an `ANTHROPIC_API_KEY` in the environment that the API
problem · Revision 1
## Observed symptom [BUG] A failed model request (e.g. HTTP 429) is reported as "the agent did not record why it stopped" When the model request fails (HTTP 429 rate limit, auth, unknown model), `autorag search` returns the "ended without
problem · Revision 1
## Observed symptom [BUG] FailoverManager.reset_all is undone by retained coordinator cooldowns SDK / package - [x] Python (`praisonaiagents`) - [ ] TypeScript / JavaScript - [ ] Cross-language / parity Environment - Provider: provider-independent; reproduction performs no networ
problem · Revision 1
## Observed symptom Static API keys get a "workos:" prefix from the client SDK and return 401 re-authenticate; re-authenticating not fixes it Summary Our client SDK (pi-multikey extension, https://www.npmjs.com/package/pi-multikey, version 1.17.0) prefixes the Authorization heade
problem · Revision 1
## Observed symptom Malformed percent-encoding in an App Router dynamic segment returns 500 instead of 400 Link to the code that reproduces this issue https://github.com/y-pakorn/next-malformed-param-500 ## Literal error text ``` returns 404, so only dynamic segments are affected
problem · Revision 1
## Observed symptom Factory Droid support to OMP Reporting a reproduction of what looks like the same issue already filed as can1357/oh-my-pi#14032, from a second and unrelated account. Two details may be new: a cross-provider comparison suggesting the fault is
problem · Revision 1
## Observed symptom bug: staging invitation targets unavailable [local] callback A user reported previously accepting a staging organization invitation and reaching an unavailable [local] destination. Current replay and separate callback checks establish the following; prior acce
problem · Revision 1
## Observed symptom Renaming a DirectusPermissionRule deletes the permission the new rule adopted `ensurePermission` (`packages/infrastructure/src/directus/client.ts:268`) adopts an existing permission row that matches (policy, collection, action) and returns its id. When a rule'
problem · Revision 1
## Observed symptom Extension resources ignore an explicit provider A resource from a kubernetes extension package ignores `options.provider` set to an explicit `kubernetes` provider and binds to the default provider instead, so it deploys to the ambient kubeconfig's cluster. pul
problem · Revision 1
## Observed symptom `pulumi convert --from yaml` panics with a nil pointer dereference on a resource with no `type` **Steps to reproduce** Versions: pulumi v3.262.0, yaml converter plugin 1.38.7, Linux arm64. ## Literal error text ``` Error: Required field 'type'
problem · Revision 1
## Observed symptom [P2] Emit only routable, unique Next.js routes Problem Next.js route detection emits `layout` files as routes and can produce duplicate route patterns when a segment contains both `layout` and `page`. The Pages Router matcher can also classify
problem · Revision 1
## Observed symptom [Bug]: GET /key/info returns 500 MissingRequiredValueError (`where.token`) when the authenticated principal has no virtual key (api_key=None) instead of a clean 4xx `GET /key/info` returns `500` with an unhandled `prisma.errors.MissingRequiredValueError: \`whe
problem · Revision 1
## Observed symptom Project-scoped commands report "generator body threw" instead of the API's error when listing projects fails `packages/cli/src/commands/project/context.ts` and `project/show.ts` pass a `listProjects` that can throw into `Result.gen` (`packages/cli/src/lib/proj
problem · Revision 1
## Observed symptom error_code: phone_provider_disabled This is a managed Supabase Cloud project, not self-hosted. Project ref: rownhfjzgcjodgoobmvt ## Environment GitHub supabase/supabase #50983 ## Trigger / context Public GitHub report in supabase/supabase (FRESH_7D). ## Source
problem · Revision 1
## Observed symptom Gemini rejects threads containing Claude thinking blocks with 400 "Invalid thought signature" (model switch, compaction, title generation) 1. In the Agent Panel, configure two providers with thinking enabled: - Claude with thinking (e.g. Claude 3.7 Sonnet /
problem · Revision 1
## Observed symptom [Bug]: Gemini errors on streaming /v1/chat/completions bypass exception mapping (raw bytes message, no ContextWindowExceededError) On `/v1/chat/completions` with `stream: true` against a `gemini/` model, an upstream error bypasses LiteLLM's exception mapping.