## Summary
Separate OSS desktop artifact publication from fleet-wide auto-update
promotion.
- retain the exact generated updater manifest as `updater-manifest.json`
on each immutable `desktop-vX.Y.Z` release
- stop the tag-triggered build from mutating
`buzz-desktop-latest/latest.json`
- add a `main`-only manual promotion workflow with one global
concurrency group
- validate stable semver, release/tag commit identity, draft/prerelease
state, exact platform set, signatures, version-bound asset URLs, asset
existence, monotonicity, idempotent retries, and a final stale-state
check before writing
- document the operator flow and pin the split with focused contract
tests
## Safety behavior
Publishing a versioned GitHub release no longer exposes it through the
in-app updater. Operators can install and test those exact
signed/notarized artifacts, then manually run **Promote OSS Desktop
Auto-Update** with the stable version.
Promotion rejects downgrades. A same-version retry succeeds only when
the rolling and candidate manifests are byte-identical. The workflow
re-reads the current rolling version immediately before its only write
and records the actor, source tag commit, previous version, manifest
digest, and run URL.
## Verification
Verified at commit `39caf1603be06bb476905225ec55f7bbbe86b237`:
```text
scripts/test-oss-desktop-promotion.sh
OSS desktop promotion contract passed
scripts/test-release-ref-contract.sh
release ref contract passed
git diff --check origin/main...HEAD
(clean)
```
The repository pre-push hook also passed `branch-skew` for the exact
pushed head; package suites were correctly skipped because this change
only touches release workflows, scripts, and documentation.
Originating conversation: Buzz channel `separate-publish-step-release`,
thread
`8857ce8bbe928e891165eddcf06c666cf6eae16181c3f02a6d8c396d8a536026`.
Signed-off-by: Wes <wesbillman@users.noreply.github.com>
Co-authored-by: Carl <c7ebe626f000404285d3686e1dc74cc07cc60a9754a150041ba132e14bd3e2ec@buzz.block.builderlab.xyz>
## Summary
Pin every GitHub CLI PR operation in the Desktop release helper to
`block/buzz`.
Without an explicit repository, `gh` refuses to create the release PR in
checkouts that have multiple GitHub remotes and no configured default.
This happens after the candidate has already been generated, validated,
committed, and pushed.
Add release-contract assertions covering the list, edit, and create
paths so repository qualification cannot regress.
## Validation
- `bash -n scripts/prepare-desktop-release.sh
scripts/test-release-ref-contract.sh`
- `scripts/test-release-ref-contract.sh`
- pre-push `branch-skew`
Signed-off-by: Wes <wesbillman@users.noreply.github.com>
Co-authored-by: Carl <c7ebe626f000404285d3686e1dc74cc07cc60a9754a150041ba132e14bd3e2ec@buzz.block.builderlab.xyz>
## Summary
- prioritize exact whole-lexeme matches within short kind-0 prefix
searches
- preserve the existing prefix result set, pagination, community/channel
scope, hydration, and authorization path
- add a Postgres regression where newer noisy `jm…` profiles saturate
the bounded page
## Why
Desktop mention autocomplete starts searching after one character. The
`jm` profile is indexed and matches both `jm:*` prefix search and
standard full-text search, but production prefix search returns a full
50-result page without it. Raw profile JSON supplies enough unrelated
`jm…` lexemes that newer equal-rank matches fill the bounded page before
the exact short display name.
Changing clients would leave deployed Desktop 0.5.8 installations
broken. This shared search-layer compatibility fix changes ordering only
for `Prefix + kinds:[0] + query length <= 2`; message search, longer
profile typeahead, and agent eligibility are untouched.
## Validation
At commit `ff88761135d5045139aeb3da14d08cbfba203169` with a clean
worktree:
- `BUZZ_TEST_DATABASE_URL=postgres://buzz:buzz_dev@localhost:5432/buzz
cargo test -p buzz-search --tests -- --include-ignored` — 22 passed (3
unit + 19 Postgres integration)
- `cargo clippy -p buzz-search --tests -- -D warnings`
- `cargo fmt --all -- --check`
- mutation check: disabling exact-lexeme priority makes
`short_kind0_prefix_prioritizes_exact_lexeme_on_a_noisy_page` fail
- mandatory pre-push hooks: branch-skew, Rust tests, and Desktop/Tauri
checks passed
## Risk
Low. The extra ordering predicate applies only to one- or two-character
prefix searches restricted exactly to kind 0. It does not add
candidates, bypass filters, or alter access control. Exact matches move
ahead of broader prefix matches; all remaining ordering stays relevance,
recency, then event ID.
Signed-off-by: Wes <wesbillman@users.noreply.github.com>
Co-authored-by: Carl <c7ebe626f000404285d3686e1dc74cc07cc60a9754a150041ba132e14bd3e2ec@buzz.block.builderlab.xyz>
## Summary
- deliver legacy ACP standing context once per live session, committing
delivery state only after a successful turn
- send only new thread/DM event deltas on later turns, with fail-open
behavior for missing IDs and failed/cancelled prompts
- fence native steer delivery acknowledgements by ACP session identity
so stale acks cannot poison replacement sessions
- keep context hints truthful when a fetch contains only the triggering
event versus history delivered earlier
## Validation
The pre-push hook passed on exact pushed head
`6a768f1bc80fe63c686acf8d730f177fff8add3c`:
- `branch-skew`
- `desktop-check`
- `desktop-typecheck`
- `desktop-test`
- `rust-tests`
- `desktop-tauri-checks`
Focused regression tests were also run while iterating:
- `channel_prompt_commits_delivery_state_only_after_acp_success`
- `in_flight_stale_native_steer_ack_cannot_update_replacement_session`
- thread/DM trigger-only versus previously-delivered context hint tests
## Known limitations and follow-ups
A local Goose smoke timed out at `session/new`. This diff does not
change code that executes at or before `session/new`; its earliest
affected runtime behavior is delivery-state insertion after session
creation succeeds. The smoke failure is therefore bounded as
environmental or pre-existing, but no successful live-provider turn was
obtained. Scripted ACP wire/lifecycle tests carry the regression
coverage.
- #5421 — distinguish post-delta, already-delivered, and fetch-truncated
context counts
- #5422 — define a standing-context re-delivery policy if a legacy
provider compacts it away
Durable process-restart/session resume remains out of scope for this
slice of #5342. #5386 also remains separate pending upstream adapter
support.
---------
Signed-off-by: Wes <wesbillman@users.noreply.github.com>
Co-authored-by: Carl <c7ebe626f000404285d3686e1dc74cc07cc60a9754a150041ba132e14bd3e2ec@buzz.block.builderlab.xyz>
## What
Implements Phases 2 and 4a of the Usage v2 plan (plan events
`d0268cd0`/`0e95b035`), extending the archive backend to emit,
transport, archive, and aggregate both cache categories and billing
identity fail-closed.
### P2 — emission, transport, archive
**Tri-state accumulators** (`Unseen`/`Exact`/`Unknown`) for cache-read
and cache-write in `buzz-agent` turn and session state. Absent field =
Unknown (never zero) through the full pipeline. No `unwrap_or(0)` on the
cache path. Both cache folds are gated on usage-bearing responses (same
gate as the total-state and identity folds) — a response with no usage
at all must not poison either accumulator.
**Overflow-aware input token parsing and accumulation** — closed
end-to-end from parse through wire to ACP:
- `sum_usage()` returns `SumUsageResult` (`Exact(u64)` | `Overflow`) —
checked arithmetic, never clamps. `anthropic_input_tokens()` returns
`Option<SumUsageResult>` since it sums three fields (`input_tokens +
cache_read_input_tokens + cache_creation_input_tokens`) that can
collectively overflow. Single-field callers (`prompt_tokens`,
`completion_tokens`, etc.) convert via `.into_exact()` — their
single-field sums cannot overflow.
- `LlmResponse.input_tokens_overflowed: bool` propagates the parse-layer
signal into the run loop. When set, `input_tokens` is `None` (clamped
value discarded), the context-gate baseline
(`last_request_input_tokens`) is frozen at its prior reading, and
`turn_input_tokens` is poisoned to `TurnIOState::Poisoned` before any
emission — including mid-turn `emit_usage_update` calls. A dedicated
enum on `LlmResponse.input_tokens` would ripple into ~20 existing test
assertions on `r.input_tokens == Some(...)`; the bool flag confines the
change to the two call sites that check it.
- `TurnIOState` (`Unseen`/`Exact`/`Poisoned`) for input and output:
per-round fold uses `checked_add`; overflow poisons permanently at turn
and session level, no healing. Absence does not poison (pass-2-cleared
contract unchanged). Wire emission omits
`accumulatedInputTokens`/`accumulatedOutputTokens` when poisoned — never
null, never `u64::MAX`. ACP treats absent = publisher-poisoned:
`delta_reliable: false`, null turn fields, null cumulative for that
category; session cumulative stays unknown for all subsequent turns once
poisoned.
**Conditional wire emission** for `accumulatedCachedInputTokens` and new
`accumulatedCacheWriteTokens`: fields are omitted when the cumulative is
Unseen or Unknown. ACP `_goose/unstable/session/update` contract
documented next to the payload with tests for all absence/zero variants.
**`PricingIdentity` stamping (publisher-side)**:
- `pricing_authority()`: canonical parsed-URL endpoint comparison
against the official allowlist — HTTPS only, exact allowlisted host
(lookalike-safe), default port (omitted or explicit :443), required API
base path, rejects userinfo/query/fragment/path-prefix lookalikes.
- Model: the actually-requested `request_model` after mesh/auto
resolution (not `effective_model_str`).
- Turn discipline: identity retained only while ALL usage in the current
turn carries one identical proven identity; any mismatch,
unproven-usage-bearing response, or unpaired cumulative snapshot poisons
to absent; a later matching notification does not heal a mixed turn.
**ACP `UsageTracker` identity fold**: per-in-flight-turn tri-state
identity accumulator replacing last-update-wins. Any absent identity on
a token-advancing notification or exact mismatch poisons to absent;
poison survives later updates; reset in `begin_turn()`/`take()`; reset
also when a request fails (baseline cleared so preflight gate cannot
stay frozen sub-threshold on retries).
**M3 migration**: adds `turn_cache_write_tokens`,
`cumulative_cache_write_tokens`, `pricing_authority`, `pricing_model`,
`pricing_cache_class` to `agent_metric_index`. Additive, idempotent,
guarded per-column by marker. M2 migration also guarded per-column (turn
and cumulative cache-read columns checked and added independently;
marker commits only after both are present). Fresh-DB schema includes
all columns.
**First-turn baselines**: `seed_zero_baseline` seeds `last_input:
Some(0)`, `last_output: Some(0)`, `last_cached_input: Some(0)`,
`last_cache_write: Some(0)`, and `last_total: Some(0)` — all have the
known-zero-at-spawn argument. Absent fields from incoming snapshots
still produce unknown (tri-state unchanged). Sessions buzz-acp did not
spawn (no seed) remain fail-closed on turn one.
**`ReportedUsage` TS mirror**: `cacheReadTokens`, `cacheWriteTokens`,
`freshInputTokens` added to `tauriArchive.ts` as `UsageField` members,
field-for-field with the Rust struct.
### P4a — aggregation layer
**Extended S-1 ladder** to cache-read and cache-write via the same
`ladder_token` path as the existing token fields.
**`freshInputTokens` derivation**: checked arithmetic, fail-closed —
absent cache fields produce Unknown (not zero), overflow and
`cacheRead+cacheWrite > input` both produce `incomplete: true`.
Aggregated as a `UsageField`.
**D6 comparator**: `sort_value()` = provider total when known, else
`input+output` when both known, else `None` (unknown-last). Replaces the
prior total-only comparator for both agent-level and model-level sort.
Ships a pinned test vector that the TS render layer (P5) must match.
## Test coverage
- `buzz-agent`: 440 lib + 15 integration (golden_transcripts) — includes
13 new `cache_total_state_tests`; 14 new `turn_io_state_tests`; 3 new
`sum_usage_*` tests (exact single-field, exact two-field, overflow
signals correctly); 3 new `parse_anthropic_*` tests (overflow flag set +
value cleared, normal sum no flag, absent usage no flag); end-to-end
golden transcript drives real subprocess with Anthropic-shaped
`input_tokens: u64::MAX, cache_read: 1` response and asserts
`accumulatedInputTokens` absent from the emitted `usage_update` — no
logic duplication; 3 wire pin tests; 4 `fold_pricing_identity_*` tests;
`pricing_authority()` explicit-:443 acceptance
- `buzz-acp`: 700 tests (691 lib + 9 integration) — 4 new usage tests
(absent input → unreliable+null; absent output → unreliable+null;
goose-shaped both present unchanged; poison mid-session); 3 ACP behavior
tests; 7 pool lifecycle tests
- Desktop (Rust): 2259+ tests — 14 new P4a pinned tests; 2 M3 round-trip
tests; 1 serde key-shape test; 2 M2 partial-schema migration tests;
first-turn cache round-trip test
## Related PRs
- P1 NIP-AM spec: [#4632](https://github.com/block/buzz/pull/4632)
- P3 pricing table: [#4629](https://github.com/block/buzz/pull/4629)
- UI (P5): [#4001](https://github.com/block/buzz/pull/4001)
---------
Signed-off-by: Will Pfleger <pfleger.will@gmail.com>
Signed-off-by: Duncan <dcfd242e557282d7a1e2cf2e6877522682f1e5c6156dc92ca7d90eaedd3b0f95@buzz.block.builderlab.xyz>
Co-authored-by: npub1mn7jgtj4w2pd0g0zeuhxsa6jy6p0rewxz4kujt98my82ahfmp72sxjexk7 <dcfd242e557282d7a1e2cf2e6877522682f1e5c6156dc92ca7d90eaedd3b0f95@buzz.block.builderlab.xyz>
## Why
Selecting or typing a member whose display name extends another member's
name, such as `@Fast Fizz Codex`, could emit p-tags for both identities
and wake the wrong agent.
## What
- Resolve overlapping member-name matches by choosing the longest valid
display name at each mention offset
- Preserve separately typed short-name mentions at different offsets
- Add regression coverage for selected team expansions and manually
typed prefix collisions
## Risk Assessment
Low to medium — this changes Desktop mention routing only. Exact
mentions and distinct offsets remain supported; same-length ambiguous
display names remain conservatively tagged because text alone cannot
disambiguate them.
Will resolve https://github.com/block/buzz/issues/2909
Generated with Goose
Signed-off-by: Hardworking Honey <c5c455215c2506cb8ba776518cec804af62d3a0526e32d496a22072e395042b9@buzz.block.builderlab.xyz>
Signed-off-by: Atish Patel <atish@squareup.com>
Co-authored-by: Hardworking Honey <c5c455215c2506cb8ba776518cec804af62d3a0526e32d496a22072e395042b9@buzz.block.builderlab.xyz>
## Summary
- temporarily allow the informational `RUSTSEC-2026-0243` advisory for
the retired `nostr-relay-pool` crate
- document the exact MeshLLM → `nostr-sdk 0.44.1` transitive path and
removal condition
- keep every other advisory and the global dependency policy enforced
## Why an exception
RustSec provides no patched `nostr-relay-pool` release because the
standalone crate was absorbed into `nostr-sdk >= 0.45`. Buzz inherits it
through pinned MeshLLM v0.74. A direct test bump to `nostr-sdk 0.45.1`
removed the retired crate but produced 13 MeshLLM API compilation
errors, so the durable fix requires an upstream source migration rather
than a lockfile update.
This narrow exception restores the required Security check while that
migration is completed. It must be removed once MeshLLM adopts
`nostr-sdk >= 0.45`.
## Validation
- `bin/cargo-deny --locked check --config deny.toml advisories`
- `bin/cargo-deny --locked check`
- `git diff --check origin/main...HEAD`
- mandatory pre-push Rust and desktop/Tauri checks
## Scope
One four-line `deny.toml` addition. No Rust source, lockfile, runtime,
or release behavior changes.
Signed-off-by: Wes <wesbillman@users.noreply.github.com>
Co-authored-by: Carl <c7ebe626f000404285d3686e1dc74cc07cc60a9754a150041ba132e14bd3e2ec@buzz.block.builderlab.xyz>
This PR contains the following updates:
| Package | Type | Update | Change |
|---|---|---|---|
| [anyhow](https://redirect.github.com/dtolnay/anyhow) | dependencies |
patch | `1.0.103` → `1.0.104` |
| [anyhow](https://redirect.github.com/dtolnay/anyhow) |
workspace.dependencies | patch | `1.0.103` → `1.0.104` |
---
> [!WARNING]
> Some dependencies could not be looked up. Check the [Dependency
Dashboard](../issues/1) for more information.
---
### Release Notes
<details>
<summary>dtolnay/anyhow (anyhow)</summary>
###
[`v1.0.104`](https://redirect.github.com/dtolnay/anyhow/releases/tag/1.0.104)
[Compare
Source](https://redirect.github.com/dtolnay/anyhow/compare/1.0.103...1.0.104)
- Update `syn` dev-dependency to version 3
</details>
---
### Configuration
📅 **Schedule**: (UTC)
- Branch creation
- Between 12:00 AM and 03:59 AM, only on Monday (`* 0-3 * * 1`)
- Automerge
- At any time (no schedule defined)
🚦 **Automerge**: Enabled.
♻ **Rebasing**: Whenever PR becomes conflicted, or you tick the
rebase/retry checkbox.
👻 **Immortal**: This PR will be recreated if closed unmerged. Get
[config
help](https://redirect.github.com/renovatebot/renovate/discussions) if
that's undesired.
---
- [ ] <!-- rebase-check -->If you want to rebase/retry this PR, check
this box
---
This PR was generated by [Mend Renovate](https://mend.io/renovate/).
View the [repository job
log](https://developer.mend.io/github/block/buzz).
<!--renovate-debug:eyJjcmVhdGVkSW5WZXIiOiI0NC4zLjIiLCJ1cGRhdGVkSW5WZXIiOiI0NC4zLjIiLCJ0YXJnZXRCcmFuY2giOiJtYWluIiwibGFiZWxzIjpbXX0=-->
Co-authored-by: renovate[bot] <29139614+renovate[bot]@users.noreply.github.com>
This PR contains the following updates:
| Package | Type | Update | Change |
|---|---|---|---|
| [arc-swap](https://redirect.github.com/vorner/arc-swap) | dependencies
| patch | `1.9.1` → `1.9.2` |
---
> [!WARNING]
> Some dependencies could not be looked up. Check the [Dependency
Dashboard](../issues/1) for more information.
---
### Release Notes
<details>
<summary>vorner/arc-swap (arc-swap)</summary>
###
[`v1.9.2`](https://redirect.github.com/vorner/arc-swap/blob/HEAD/CHANGELOG.md#192)
- Document RefCnt must not panic
([#​208](https://redirect.github.com/vorner/arc-swap/issues/208)).
</details>
---
### Configuration
📅 **Schedule**: (UTC)
- Branch creation
- Between 12:00 AM and 03:59 AM, only on Monday (`* 0-3 * * 1`)
- Automerge
- At any time (no schedule defined)
🚦 **Automerge**: Enabled.
♻ **Rebasing**: Whenever PR becomes conflicted, or you tick the
rebase/retry checkbox.
👻 **Immortal**: This PR will be recreated if closed unmerged. Get
[config
help](https://redirect.github.com/renovatebot/renovate/discussions) if
that's undesired.
---
- [ ] <!-- rebase-check -->If you want to rebase/retry this PR, check
this box
---
This PR was generated by [Mend Renovate](https://mend.io/renovate/).
View the [repository job
log](https://developer.mend.io/github/block/buzz).
<!--renovate-debug:eyJjcmVhdGVkSW5WZXIiOiI0NC4zLjIiLCJ1cGRhdGVkSW5WZXIiOiI0NC4zLjIiLCJ0YXJnZXRCcmFuY2giOiJtYWluIiwibGFiZWxzIjpbXX0=-->
Co-authored-by: renovate[bot] <29139614+renovate[bot]@users.noreply.github.com>
This PR contains the following updates:
| Package | Type | Update | Change |
|---|---|---|---|
| [diffy](https://redirect.github.com/bmwill/diffy) | dependencies |
patch | `0.5.0` → `0.5.1` |
---
> [!WARNING]
> Some dependencies could not be looked up. Check the [Dependency
Dashboard](../issues/1) for more information.
---
### Release Notes
<details>
<summary>bmwill/diffy (diffy)</summary>
###
[`v0.5.1`](https://redirect.github.com/bmwill/diffy/blob/HEAD/CHANGELOG.md#051---2026-07-18)
[Compare
Source](https://redirect.github.com/bmwill/diffy/compare/0.5.0...0.5.1)
##### Fixed
- [#​85](https://redirect.github.com/bmwill/diffy/pull/85)
Merge conflict markers are now always placed on their own lines.
Previously, a conflicting hunk at the end of a file without a trailing
newline glued the next marker onto its last content line, producing
unparseable output. This matches `git merge-file --diff3` behavior.
</details>
---
### Configuration
📅 **Schedule**: (UTC)
- Branch creation
- Between 12:00 AM and 03:59 AM, only on Monday (`* 0-3 * * 1`)
- Automerge
- At any time (no schedule defined)
🚦 **Automerge**: Enabled.
♻ **Rebasing**: Whenever PR becomes conflicted, or you tick the
rebase/retry checkbox.
👻 **Immortal**: This PR will be recreated if closed unmerged. Get
[config
help](https://redirect.github.com/renovatebot/renovate/discussions) if
that's undesired.
---
- [ ] <!-- rebase-check -->If you want to rebase/retry this PR, check
this box
---
This PR was generated by [Mend Renovate](https://mend.io/renovate/).
View the [repository job
log](https://developer.mend.io/github/block/buzz).
<!--renovate-debug:eyJjcmVhdGVkSW5WZXIiOiI0NC4zLjIiLCJ1cGRhdGVkSW5WZXIiOiI0NC4zLjIiLCJ0YXJnZXRCcmFuY2giOiJtYWluIiwibGFiZWxzIjpbXX0=-->
Co-authored-by: renovate[bot] <29139614+renovate[bot]@users.noreply.github.com>
## Summary
- remove the complete Welcome guidance surface when dismissal reaches
`hidden`
- preserve dismissal across the private and starter Welcome channels for
the active identity
- assert the starter channel's actual `welcome-everyone` title on
re-entry
## Why
PR #5330 introduced two deterministic Desktop E2E failures:
- the inner banner unmounted, but `welcome-composer-guidance-layer`
remained
- the re-entry test expected case-sensitive `Welcome` while navigating
to `welcome-everyone`
The state hook also scoped completion to channel IDs while `ChannelPane`
remounts during navigation. The Welcome guidance is one experience
spanning both Welcome channels, so completion now survives that remount
while remaining identity-scoped.
## Validation
At `b577eb42edffe889f63566f2457eacea720f3593`:
- `pnpm -C desktop typecheck`
- focused Biome check for all four changed files
- E2E build
- both `welcome-everywhere banner` integration tests repeated three
times: **6/6 passed**
- mandatory pre-push desktop check, typecheck, and full desktop unit
suite: **4,535 passed**
- `git diff --check`
Signed-off-by: Wes <wesbillman@users.noreply.github.com>
Co-authored-by: Carl <c7ebe626f000404285d3686e1dc74cc07cc60a9754a150041ba132e14bd3e2ec@buzz.block.builderlab.xyz>
## Problem
A provider can return HTTP 200 with a **truncated JSON body** — cleanly
closed connection, correct framing, content cut off mid-value. Both LLM
HTTP loops treated this as a terminal error on the first attempt:
`AgentError::Llm("json: EOF while parsing a value")`, surfaced as code
-32000 at the ACP boundary, killing the agent turn before it produced
anything.
Observed live in a tb2.1 bench trial (write-compressor, tb21-twins-1):
deepseek via OpenRouter returned a truncated body, the agent died
mid-prompt with 0 turns completed, and the trial scored 0 on a provider
hiccup.
Meanwhile the same loops already retry timeouts, 429s, 5xxs, 499s, and
mid-body stream stalls — a truncated-but-complete body was the one
transient upstream fault that fell through to terminal.
## Fix
In both `post()` and `openrouter_post()`
(`crates/buzz-agent/src/llm.rs`): when the fully-received success body
fails `serde_json::from_slice`, `continue` the **existing** retry loop
instead of returning terminal — same `MAX_RETRIES` (3) bound, same
`backoff_with_jitter`. On exhaustion, the error goes through
`terminal_llm_error` so it carries cumulative duration + attempt count
like every other retried failure (previously the `json:` error carried
neither).
`post_anthropic` routes through `post()`, so
Anthropic/OpenAI/Databricks/mesh and OpenRouter are all covered.
## Why this cannot re-run a tool call
Hard requirement: tool calls are not idempotent, and this change must
not introduce any possibility of replaying one.
1. **The retry lives inside the HTTP POST helper, below the parse
boundary.** Tool calls are only ever extracted from a *successfully
parsed* response value
(`parse_openai`/`parse_anthropic`/`parse_responses`, all downstream of
these helpers' `Ok` return). A malformed body never parses, therefore no
tool call was ever extracted from it, therefore nothing downstream of it
ever dispatched.
2. **What is re-sent is the completion request itself** — the identical
`body_bytes` captured once at function entry. Sending a completion
request executes no tools; it asks the model for the next message.
3. **Same safety class as existing behavior.** The loop already re-sends
this identical request on 429/5xx/timeout/stream-stall; this adds one
more transient-fault arm to the same loop with the same bytes.
## Tests
Three new tests mirroring the existing 499/dropped-connection fixtures
(raw `TcpListener` stubs):
- `post_retries_malformed_json_body_and_succeeds` — truncated 200 body
on attempt 1, valid JSON on attempt 2; asserts success and **exactly 2**
server-side requests
- `post_exhausts_retries_on_persistent_malformed_json` —
always-truncated body; asserts exactly `MAX_RETRIES` attempts and a
terminal error carrying `json:` + cumulative/attempt context
- `openrouter_post_retries_malformed_json_body_and_succeeds` — same
recovery through OpenRouter's separate loop
Full `cargo test -p buzz-agent` green at e7a5d7bb (430 lib + all
integration targets, 0 failures); `cargo fmt` + `clippy --all-targets`
clean.
Originating conversation: buzz-benchmarking channel, thread 397a992d.
Signed-off-by: Eva <011987e296fd5006292d2f930b574be47c7801048d1983c46c425d3c95f0cffd@buzz.block.builderlab.xyz>
Co-authored-by: Eva <011987e296fd5006292d2f930b574be47c7801048d1983c46c425d3c95f0cffd@buzz.block.builderlab.xyz>
## Problem
The `WelcomeComposerGuidanceLayer` in the `#Welcome` channel was
positioned with `absolute inset-x-0 bottom-full z-[-1]` — outside the
`composerWrapperRef` measurement boundary. `useComposerHeightPadding`
observes `composerWrapperRef`'s block size to set `paddingBottom` on the
timeline scroll container, but the absolutely-positioned layer didn't
contribute to that size. The banner sat directly on top of the newest
message, blocking the thread affordance on that message, and had no
manual dismiss control.
## Fix
**Overlap**: Changed `WelcomeComposerGuidanceLayer` from `absolute
inset-x-0 bottom-full z-[-1]` to `relative` (in normal flow). As a
normal-flow child of `composer-dock`, the layer's full height is now
measured by the ResizeObserver and fed into the timeline's
`paddingBottom`, so the newest message is always fully visible and its
thread affordance is always clickable while the banner shows.
**Dismiss**: Added an `X` close button
(`data-testid="welcome-composer-dismiss-button"`) on the prompt state.
Clicking fires `onDismiss`, which drives `dismissing → hidden`
immediately (same slide-down animation as the auto-dismiss path) and
marks the channel ID as completed in the session ref so the banner does
not reappear on channel re-entry within the session.
**Refactor**: Extracted the banner state machine (refs, timers,
`useEffect`s, and callbacks) from `ChannelPane.tsx` into
`useWelcomeComposerBanner.ts`. This keeps `ChannelPane.tsx` well under
the 1000-line file-size ratchet and makes the state machine
independently testable.
## Changed files
- `desktop/src/features/channels/ui/WelcomeComposerBanner.tsx` —
`WelcomeComposerGuidanceLayer` positioning fix; `onDismiss` prop;
dismiss button; `overflow-hidden` / `mb-0` / `flex-1` cleanup
- `desktop/src/features/channels/ui/ChannelPane.tsx` — remove inline
banner state machine, use `useWelcomeComposerBanner` hook, pass
`onDismiss`
- `desktop/src/features/channels/ui/useWelcomeComposerBanner.ts` — new
hook owning all banner state
---------
Signed-off-by: Will Pfleger <pfleger.will@gmail.com>
Co-authored-by: Hayt <41ea58f1e64c243627e8acde7c89be667052ee6e17d8f021c1195be4324ebf04@buzz.block.builderlab.xyz>
The Prompt Context modal (observer feed → check icon under sent
messages) was clipping all content and card right-padding at the dialog
edge.
**Root cause**: `PromptContextDialog` renders inside `DialogContent`,
which is a CSS grid. The child flex wrapper had default `min-width:
auto`, so the widest unbreakable token in the content (64-char hex event
IDs, `Tags: [[...]]` JSON) set the grid track width, blowing it past
`max-w-xl`. `overflow-hidden` then clipped everything at the dialog edge
— including the section cards' right padding.
**Fix**:
- `AgentSessionTranscriptList.tsx`: add `min-w-0` to the `flex
max-h-[85vh] flex-col` wrapper so the grid item can shrink below its
max-content width.
- `PromptSectionAccordion.tsx`: replace `wrap-break-word` with
`wrap-anywhere` on the body text (open and collapsed states) and the
title. `overflow-wrap: anywhere` reduces min-content width, which
`break-word` does not, letting long tokens wrap inside the cards rather
than inflating the track.
The `line-clamp-2` collapsed preview is preserved unchanged.
Signed-off-by: Will Pfleger <pfleger.will@gmail.com>
Co-authored-by: Duncan <dcfd242e557282d7a1e2cf2e6877522682f1e5c6156dc92ca7d90eaedd3b0f95@buzz.block.builderlab.xyz>
## Buzz Desktop release v0.5.8
- **Frozen main:** `6a17d035f79ad582ca3f4f3cdc38d376f2c4087f`
- **Reviewed candidate:** `f3de860574bb3119018b4592353e9761635aeb07`
- **Previous desktop release:** `desktop-v0.5.7`
- **Proposed immutable tag:** `desktop-v0.5.8`
This PR may be **squash merged** after the Desktop Release Candidate
check and all protected-branch checks pass. Merging authorizes
publication of the exact reviewed candidate; later or unrelated changes
on `main` cannot alter it.
The checked-in changelog accounts for every non-merge commit in the
release range. The Desktop tag points to the reviewed candidate commit,
not the later squash commit. Publication remains bound to that immutable
candidate tag.
Signed-off-by: Wes <wesbillman@users.noreply.github.com>
Co-authored-by: Release Automation <release-automation@users.noreply.github.com>
