## Observed symptom [Bug]: CometAPI issue while invoking Claude Opus 5.5 (thinking) The result of my last try session even before the model answer : ── Hermes error details ── time: 2026-09-25T13:12:30.575Z layer: provider code: reasoning_mandatory retryable: true provider: custo
## Observed symptom Bedrock: Claude Opus 5.5 returns 400 in from_bedrock TOOLS mode because toolChoice is always forced Thanks for #2566 and the other recent Bedrock work that made the handlers keep caller-owned nested config. This looks like one more
## Observed symptom [BUG] Bedrock: Claude Opus 5.5 structured output fails with 400 because toolChoice is forced Thanks for the Bedrock structured-output support from #4280 and the "append `structured_output` without forcing" branch from #4307. With Claude Opus 5.5 on Bedrock,
## Observed symptom Deno JSR packages are never found in a real Deno project: the crawler only probes $DENO_DIR/npm/jsr.io, and the matching ./vendor/jsr.io layout from "vendor": true is ignored [agent] Found by the scheduled Deno bug-hunt routine (ledger #308). Summary
## Observed symptom - [x] I'm reporting a bug unrelated to a specific site. - [x] I've verified that I'm running the [**latest version**](https://github.com/yt-dlp/yt-dlp/releases/latest) of yt-dlp. - [x] I've verified that I'm running the latest [**stable version**](https://gith
## Observed symptom Global mode can't see any Deno global install: the npm packages under $DENO_DIR/npm/registry.npmjs.org are never crawled, so scan -g misses them and Deno 2.9's per-tool node_modules gets patched and VEX-attested while the tool runs the unpatched copy
## Observed symptom [Bug]: `deno deploy` CLI appends an extra trailing newline to deno.json on every run with --org Every authenticated run of the Deno Deploy CLI (`@deno/deploy`, used by `deno deploy`) with `--org` rewrites `deno.json` and adds one more
## Observed symptom `bun test --parallel`: a worker that panics after its last test file does not fail the run What version of Bun is running? 1.4.3-canary.1 (main at 2722608f47). The files named below are the same at 5a183c1ebc. ##
## Observed symptom [Segfault at 0x10 in nextTick drain on DarkWake after 5.6 days — 2.1.278, Bun 1.4.3 (8c28b3167), macOS 27.0 arm64] - [x] I have searched [existing issues](https://github.com/anthropics/claude-code/issues?q=is%3Aissue%20state%3Aopen%20label%3Abug) and this hasn
## Observed symptom [BUG] Bun 1.4.3 segfault (address 0x0) mid-session in long-running claude -p on Linux x64 (seen on 2.1.284 and 2.1.286) ---- Comment by Claude Opus 5.5 ---- Long-running headless sessions (`claude -p`) crash partway through with a Bun
## Observed symptom [BUG] 2.1.284 regression: Bun 1.4.3 panics on startup on Linux x64 (`index out of bounds: the len is 0 but the index is 3`); 2.1.283 works After auto-updating to **2.1.284**, `claude` crashes on launch with a Bun
## Observed symptom [BUG] Variable Library metadata request failed in UDF after deployment. Authentication Method Hi, I'm deploying a User Data Function connected to a variable library. The deployment occurs in GitLab pipelines with Service Principal owner of the target
## Observed symptom 🐛 azure: --tenant-id is ignored with Azure CLI authentication; the token comes from the CLI's default tenant With Azure CLI authentication, the azure provider ignores `--tenant-id`. It always requests the token from the Azure CLI's **default** tenant,
## Observed symptom Could not associate Load Balancer in VMSS when using az cli 2.87.0+ versions We are observing an issue after upgrading Azure CLI from 2.75.0 to 2.87.0 and higher versions. Our Azure DevOps BlueGreen deployment pipeline performs traffic
## Observed symptom Protected-paths list has drifted across its parallel copies again: .gitlab-ci.yml missing from enforcement, .gitignore missing from fix-agent docs PR #1355 added `.gitignore` to `REVIEW_PROTECTED_PATHS` in both `env.runner` and `env.sandbox` blocks of `harness
## Observed symptom fix(score): affectedDays counts only an incident's start day — a 3-day outage is 1 affected day, unlike every provider page and AIWatch's own calendar `affectedDays30d` counts **one day per incident: the UTC day it started** (`incidentDay`, `worker/src/utils.t
## Observed symptom [BUG] HTTP MCP: OAuth succeeds, then initialize hangs until CONNECT_TIMEOUT (raising MCP_TIMEOUT doesn't help) - [x] I have searched existing issues and this hasn't been reported yet - [x] This is a single bug report - [x]
## Observed symptom [Bug] Anthropic API Error: Auto mode classifier unavailable blocks Bash/Edit calls **Bug Description** Bug — Auto mode classifier unavailable (server-side) blocks all Bash and Edit calls; read-only tools still work · area: permissions / auto mode ▎