## Summary
Bump the pinned Rust toolchain from 1.95.0 to 1.98.1 so release builds
work on macOS 27 hosts.
Rust ≤ 1.97 strips Apple-target artifacts with its bundled
`llvm-objcopy`, which leaves the Mach-O `LC_SYMTAB` string pool
(`__LINKEDIT`) only 4-byte aligned. macOS 27's dyld refuses to `dlopen`
such images (`mis-aligned LINKEDIT string pool`). Because cargo applies
the release profile's default `strip = "debuginfo"` to proc-macro dylibs
too, `cargo build --release` of any crate that uses sqlx macros
(`buzz-db`, `buzz-relay`, `buzz-admin`, …) fails on macOS 27: rustc
cannot load `libsqlx_macros-*.dylib`. Dev builds pass because the dev
profile does not strip, which is why this only shows up as a
release/packaging failure.
The fix (llvm/llvm-project#203680, pulled into rustc by
rust-lang/rust#158410 for rust-lang/rust#157750) first shipped in Rust
1.98.0 and was never backported to 1.97.x. `block/buzz-app` pinned
1.98.1 for the same failure in block/buzz-app#529. With this bump the
`CARGO_PROFILE_RELEASE_BUILD_OVERRIDE_STRIP=none` / custom-linker
workarounds people have been using locally are no longer needed.
### What changes
Two commits: the toolchain bump (with the desktop clippy fixes), then
the workspace clippy fixes.
| Location | Change |
|---|---|
| `rust-toolchain.toml` | `channel = "1.95.0"` → `"1.98.1"` (profile
unchanged) |
| `Dockerfile`, `Dockerfile.push-gateway` | `ARG RUST_VERSION=1.95` →
`1.98` |
| `Dockerfile.sprig` | `rust:1.98-alpine3.22` with the manifest-list
digest verified on Docker Hub (shared by the `1.98` and `1.98.1` tags) |
| `.github/workflows/release.yml` | `dtolnay/rust-toolchain` pinned to
the tip of its `1.98.1` branch (`18f04874`) for the Windows desktop
release job |
| `benchmarks/harbor-buzz-orchestra` | `RUST_IMAGE = rust:1.98-alpine` |
| `_ci-rust.yml`, `windows-canary.yml`, `app_state_tests.rs` | stale
"1.95" comments reworded; the `unknown_lints` allow stays with a
version-agnostic rationale |
| `desktop/src-tauri` | lints clippy 1.98 newly raises under `-D
warnings`: `chunks_exact_to_as_chunks` at five fixed-width PCM/UTF-16
decoders, `needless_borrows_for_generic_args` on the `Copy` `norm`
closure in `config_bridge/reader.rs`, and a redundant `Engine as _`
import in `hpke_key_backup_tests.rs` that `use super::*` already
provides |
| Workspace crates (9) | the 30 lints clippy 1.98 newly raises in `cargo
clippy --workspace --all-targets -- -D warnings`, listed below |
Workspace lint fixes:
| Lint | Sites | Fix |
|---|---|---|
| `result_large_err` (20) | `buzz-relay` `api/git/transport.rs` (8),
`api/git/settings.rs` (6), `api/gifs.rs`, `api/workflows.rs`,
`api/profiles.rs`; `buzz-push-gateway` `verify_installation_assertion` |
Clippy 1.98 now checks `async fn`s, so every `async fn … -> Result<_,
axum::response::Response>` trips it. These follow the existing allow
pattern in `api/media.rs` and `nip_fi_http.rs` ("Response is the natural
error type for axum handlers"): module-level in the two git files,
per-function elsewhere. Boxing `Response` would touch every handler for
no runtime benefit. |
| `chunks_exact_to_as_chunks` (3) | `buzz-media` MP4 brand scan;
`buzz-voice` npy f32 reader and stereo test fixture | `chunks_exact(N)`
→ `as_chunks::<N>().0` |
| `question_mark` (3) | `buzz-acp` queue flush fallback and
slash-command mention skip; `buzz-agent` `find_git_root` | `match … {
None => return None }` → `?` |
| `useless_format` (2) | `buzz-auth` nip_fi startup test policies |
`format!("…")` → `"…".to_string()` |
| `useless_borrows_in_formatting` (1) | `buzz-pairing-cli` secret print
| drop the borrow |
| `manual_noop_waker` (1) | `buzz-sdk` broker test `block_on` |
`Waker::noop()` (stable since 1.85) |
All `as_chunks` rewrites keep `chunks_exact`'s remainder-dropping
semantics, and `as_chunks` is stable since 1.88, which is the MSRV.
One change in `api/git/transport.rs` is not a lint fix. Staging that
file trips the pre-commit secret scanner on its existing test-only ES256
fixture (`GIT_TEST_EC_PEM`), and the scanner only honours a suppression
marker on the flagged line. So the `\`-continued string literal is
rewritten as a byte-identical `concat!` whose `BEGIN` line carries
`sadscan:disable kingfisher.privkey.2`.
Left as is on purpose: `Cargo.toml` `rust-version = "1.88.0"` (MSRV
floor, a separate concern), the README/CONTRIBUTING "1.88+" text, Hermit
(no Rust packages there), the lockfiles, and the `(1.95.0)` comment in
`scripts/start-isolated-test-relay.sh` (the pre-commit secret scanner
flags that file's pre-existing dev-only `DATABASE_URL` whenever it is
staged).
### Related issue
Fixes#8074
### Testing
All under rustc/clippy 1.98.1 on a macOS 27 host (Darwin 27.0.0,
aarch64), with no `RUSTFLAGS` / `SDKROOT` / strip overrides in the
environment:
- `cargo build --release -p buzz-db` (the #8074 repro: the release
profile strips the proc-macro dylib): **builds**. The resulting
`libsqlx_macros-*.dylib` (14 MB) has `LC_SYMTAB stroff % 8 == 0` per
`otool -l`. The same build on 1.95.0 died with `mis-aligned LINKEDIT
string pool`.
- `just clippy` (workspace, `-D warnings`) and `cargo fmt --check`
(workspace and `desktop/src-tauri`): pass
- `just desktop-tauri-clippy` (default and `mesh-llm` feature sets):
pass
- `just test-unit`: every crate passes. The `buzz-acp` binary failed
only under pre-push lane contention; rerun solo it is 1060 passed, 0
failed.
- `just desktop-tauri-test`: 3422 passed, 0 failed
- After the lint fixes: lib/bin unit tests for `buzz-acp`, `buzz-agent`,
`buzz-auth`, `buzz-media`, `buzz-voice`, `buzz-sdk`, `buzz-pairing-cli`,
and `buzz-relay`'s `api::transport` tests, including the
Postgres-backed
`enforce_mode_git_pack_routes_missing_assertion_exact_bytes`, which
parses the rewritten PEM: pass
`cfg(windows)` code was not linted locally (no Windows target
installed). CI's Windows Rust job runs workspace clippy for
`x86_64-pc-windows-msvc`, and that passes on this branch. CI also builds
the relay, push-gateway and sprig images on the new `rust:1.98` bases
for amd64 and arm64.
No runtime behavior changes.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
---------
Signed-off-by: Matt Toohey <contact@matttoohey.com>
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
## Why
Evaluating buzz agent memory retrieval by seeding a memory then asking
the buzz agent a question it needs that memory.
**Bug Found**: System prompt had no inclusion of retrieving cold
memories and suggested looking in a mem/*.md directory that does not
exist. Updated `system-prompt.md` to include memory CLI tools and usage.
Eval Before System Prompt Change: 0/3
Eval After System Prompt Change: 3/3
## What
- Add a `memory-retrieval` benchmark that seeds agent memory with `buzz
mem set` before asking a direct question.
- Grade the observable threaded answer without inspecting tool calls or
exposing the answer in channel history.
- Teach agents to use `buzz mem set`, `buzz mem ls`, and `buzz mem get`
for cold memory.
- Add a wire-debug endpoint configuration for diagnosing ACP tool calls
in local runs.
- Add fixture, seeding, verifier, and prompt coverage.
## Risk Assessment
Low. The runtime changes are limited to the benchmark harness. The
production-facing change clarifies existing memory commands in the base
prompt; it does not change memory storage, relay behavior, or
authorization.
## References
- Before the system-prompt changes, 0/3 attempts passed because agents
never invoked the `buzz mem` CLI and instead searched a non existent
filesystem
- After the changes, 3/3 attempts passed. ACP wire logs confirmed that
every agent ran `buzz mem ls` followed by `buzz mem get` and returned
`net_gpv`.
---------
Signed-off-by: Philip Azar <pazar@squareup.com>
## Summary
- classify every Buzz-native benchmark task as `regression` or
`workflow`
- add metadata-backed `--layer` selection with k=1 regression and k=3
workflow defaults, while preserving explicit `--attempts`/`-k`
- split an unqualified mixed Buzz dataset run into two Harbor jobs so
each layer gets the correct default
- document the layer contract and cover metadata, selection, defaults,
overrides, invalid input, and command construction
---------
Signed-off-by: Salman Mohammed <smohammed@squareup.com>
## Why
The Buzz-native runtime could stop a solo agent after one turn even when
a scripted follow-up event was queued or already running. That could
turn a harness timing race into a benchmark failure.
Follow-up to #6448.
## What
- Treat solo tasks with scripted events separately from ordinary
one-turn tasks.
- Wait until turn counts and authored message IDs remain idle and
unchanged for several polls before stopping the agent.
- Remove the message-count shortcut that could stop an active turn.
- Add focused regressions for a delayed follow-up turn and an
already-running turn with a `DONE:` message.
## Testing
- `pytest -q tests/test_container_runtime.py -k "solo_turn_end or
scripted_events"` — 3 passed
- `ruff check` on the three touched Python files
- `ruff format --check` on the three touched Python files
- Local `buzz-native-solo-luna.yaml` Harbor run, one attempt per
affected task:
- `cross-thread-requests` — reward 1.0; all 6 task-specific checks
passed
- `interleaved-agent-reports` — reward 1.0; all 7 task-specific checks
passed
- 2 completed trials, 0 exceptions, 56 seconds total
## Risk
Low. This only changes completion detection for solo benchmark tasks
with scripted events. Ordinary solo tasks retain immediate completion
after their first turn ends.
---
**Update Aug 21, 13:07 EDT:** Replaced the fixed settle delay with exact
delivery receipts after [review
feedback](https://github.com/block/buzz/pull/6487#discussion_r3832050452).
- ACP now records the event IDs delivered by each completed turn; the
runtime also recognizes existing successful-steer receipts.
- Scripted trials stop only after every expected event ID is
acknowledged and no turn is active. Missing receipts wait for the trial
budget instead of producing partial evidence.
- TDD regression delayed delivery beyond the old five-poll boundary and
failed before the implementation; the receipt parser and Rust receipt
format are pinned independently.
- Focused verification: 31 container-runtime tests passed, 6 ACP
delivery tests passed, plus Ruff, rustfmt, and Clippy.
- Receipt-gated Luna run: 2 completed trials and 0 exceptions.
`interleaved-agent-reports` scored 1.0. `cross-thread-requests`
completed both calculations and thread isolation but Luna omitted the
user mention on ALPHA, so that model-output dimension scored 0.
- Risk remains low: the ACP production change adds delivery receipt
logging and restores a missing turn-end log on a
completed-before-control race; it does not change queue dispatch
behavior.
Generated with Codex
---------
Signed-off-by: Salman Mohammed <smohammed@squareup.com>
## Why
Terminal task correctness alone does not guard Buzz-native collaboration
behavior such as exact reply routing, non-waking narrative names,
batched agent synthesis, thread isolation, and ambiguous identity
targeting.
## What
- Add five medium/hard Harbor tasks for multiline delivery, narrative
agent names, interleaved agent reports, cross-thread requests, and
ambiguous user mentions
- Add signed scripted-event fixtures, duplicate-display-name profiles,
and public-only evidence export support
- Add positive and adversarial verifier fixtures covering dropped
inputs, extra posts, incorrect routing, and unintended mentions
## Risk Assessment
Low — changes are limited to benchmark tooling and datasets; production
behavior is exercised but not modified.
## References
- Grounded in behavior reported in `buzz-community` and public issues
#5787, #5176, #5839, #4942, #4072, #4303, and #6257
- Live Claude Sonnet 4.6 smoke run: 2/5 passed; the three failures
exposed top-level instead of threaded delivery, empty narrative output,
and silent ambiguous-identity completion
- Live GPT-5.6 Luna run (`buzz-native-solo-luna.yaml`): 3/5 passed.
Passed `ambiguous-user-mention`, `cross-thread-requests`, and
`interleaved-agent-reports`. Failed `multiline-message` and
`narrative-agent-names` because both omitted the event-level user
mention despite otherwise-correct content and threading.
<img width="1176" height="805" alt="Screenshot 2026-08-21 at 11 11
17 AM"
src="https://github.com/user-attachments/assets/fc259ec9-892c-4436-b022-1bf85415617d"
/>
Signed-off-by: Salman Mohammed <smohammed@squareup.com>
## Why
Terminal-Bench can show task correctness while missing Buzz-specific
collaboration failures, such as posting a final answer outside the
user's thread. Add a small Buzz-native suite that scores relay behavior
directly.
## What
- Add Harbor tasks for direct thread replies, event-level callback
mentions to a three-word user identity, targeted reads of explicitly
named paths outside the workspace, and exact temporary-channel
membership against a seeded directory of 50 users and 10 bots
- Export sanitized post-agent relay evidence and extend solo
runtime/provisioning for deterministic verification
- Expose channel TTL in CLI search output and cover positive and
negative transcript/state fixtures
## Risk Assessment
Low to medium — this primarily adds benchmark and testbed machinery. The
CLI change is an additive `ttl_seconds` field in channel-search JSON;
relay behavior is unchanged.
## References
- Motivated by the Buzz-specific behavior observed while benchmarking
https://github.com/block/buzz/pull/6128
- The named-path regression case covers the issue being fixed by
https://github.com/block/buzz/pull/6261
- Local live Sonnet runs: `reply-to-thread`, `user-mention`, and
`create-channel-invite-users` each scored reward 1.0 with zero
exceptions
- The current branch prompt scored 3/5 on a concurrent
`read-named-path-outside-workspace` batch: one explicit
workspace-boundary refusal and one missing final response
Generated with Codex
---------
Signed-off-by: Salman Mohammed <smohammed@squareup.com>
## Problem
Two failure modes from the `tb21-glm52-crusoe-1` benchmark run (GLM-5.2
solo, TB2.1) wedged or killed 13 of 89 trials without the model being at
fault:
1. **Conversation poisoning on text-only endpoints.** Crusoe's
serverless `crusoeai/GLM-5.2-NVFP4` rejects any request whose history
contains an image with `400: ... is not a multimodal model`. The
recovery machinery for exactly this case already exists —
`AgentError::UnsupportedImageInput` → `replace_unsupported_images()`
strips the image blocks, marks the tool result as an error, substitutes
a text placeholder, and continues the turn. But classification only
matched OpenRouter's 404 body (`no endpoints found that support image
input`) and was only consulted on the 404 arms. The Crusoe 400 fell
through to terminal `AgentError::Llm`: the image stayed in history,
every subsequent call failed identically, buzz-acp rode its 10-retry
ladder (~40 min), and the trial idled to budget death. Measured blast
radius: **8 trials wedged, 12.7h aggregate idle-after-poison.**
2. **Bounded agent rounds in benchmark trials.** The harness default
`DEFAULT_MAX_AGENT_ROUNDS = 32` ended solo trials mid-work when turns
rotated (thinking-heavy models hit max_tokens rotation fast; 4 trials
died this way). Benchmark trials already have a wall-clock budget as the
real limit — the round cap only converts recoverable rotation into trial
death.
## Fix
- `is_unsupported_image_input_error()` also matches the verbatim `is not
a multimodal model` body. Matcher stays deliberately tight (same
doctrine as `is_context_length_error`): misclassifying a generic 400 as
recoverable would mutate history for an error that removing images
cannot fix.
- Both status ladders — shared `post()` and `openrouter_post()` —
consult it on their 400 arms and return the typed
`UnsupportedImageInput` (OpenAI-compatible providers report this as 400;
a BYOK/passthrough upstream can surface the provider's own 400 through
OpenRouter).
- Harness `DEFAULT_MAX_AGENT_ROUNDS` → `0` (unbounded —
`BUZZ_AGENT_MAX_ROUNDS=0` is the agent config's documented unbounded
value). Per-agent `budget.max_calls` in manifests still overrides.
## Acceptance
- A 400 with the image-rejection body reaches the existing image-strip
recovery path instead of wedging the session — asserted through
`complete()` (covers the return path into the convergence mapper) and at
the `openrouter_post` terminal, both proving single-attempt (a
deterministic capability rejection must never be retried).
- Ordinary 400s stay terminal `AgentError::Llm` (existing negative tests
unchanged).
- Benchmark trials run unbounded rounds by default; python tests updated
for 0-is-legal with a negative arm at -1.
## Verification
- `cargo test -p buzz-agent`: 427 + 18 + 20 + 15 + 8 + 1 + 48 passed, 0
failed (full package, 3 consecutive clean runs)
- `cargo clippy -p buzz-agent --all-targets`, `cargo fmt --check`: clean
- `uv run --extra dev pytest tests/` in harbor-buzz-orchestra: 35 passed
- Pre-push hooks (full workspace rust-tests + desktop-tauri-checks)
green on rustc 1.95.0 at head b0438602
Signed-off-by: Eva <011987e296fd5006292d2f930b574be47c7801048d1983c46c425d3c95f0cffd@buzz.block.builderlab.xyz>
Co-authored-by: Eva <011987e296fd5006292d2f930b574be47c7801048d1983c46c425d3c95f0cffd@buzz.block.builderlab.xyz>
## Problem
The harness posts each trial's task via `buzz messages send`, relying on
`@<orchestrator-id>` name resolution. Task text is untrusted payload:
when it contains @-tokens of its own, the CLI's mention resolver tries
to resolve them as channel members, fails, and refuses to send — killing
the trial with `RuntimeLaunchError` before the agent ever saw the task.
Live occurrence: TB 2.1's `large-scale-text-editing` task embeds Vim
macros (`:%normal! @a`). In the tb21-solo-1 run the trial died at
launch:
```
RuntimeLaunchError: buzz messages send ... exited 1:
{"error":"user_error","message":"mention '@a' does not match a current channel member; retry with --mention <pubkey>"}
```
Any TB task whose statement contains @-syntax is silently zeroed this
way.
## Fix
Pass the orchestrator's pubkey as an explicit `--mention` when posting
the task. The CLI demotes unresolved @-tokens in the text to
presentation-only when any explicit identity is supplied, so delivery
still targets exactly the orchestrator and every @-token in the task
statement becomes inert. The harness already holds the orchestrator's
`AgentCredential` (it writes that pubkey into the worker roster tables),
so no persistence is needed — fresh key per trial, fresh `--mention` per
trial.
Verified both halves against a live relay: a fenced `@a` without
`--mention` still hard-fails (the resolver is not markdown-aware); the
same content with an explicit `--mention` sends clean with
`mention_pubkeys` containing only the target.
## Testing
- `benchmarks/harbor-buzz-orchestra`: full pytest suite — 35 passed (34
baseline + new `test_send_mentions_by_pubkey_so_task_text_stays_inert`),
ruff clean. Run against `origin/main` 769ac70b with exactly this patch
applied.
- `testbed`: full pytest suite — 23 passed, 1 skipped; ruff clean.
## Acceptance
A task statement containing arbitrary @-tokens (Vim registers, emails,
decorators) launches and delivers to the orchestrator instead of dying
in `_send`.
Originating Buzz thread:
`buzz://message?channel=c3252dd2-0142-4e01-88c7-a2183c3960a5&id=74a65a0990fd2197882b66b5ea2707169d4a3dbd2020d1610c45150fb99f140b`
Signed-off-by: Eva <011987e296fd5006292d2f930b574be47c7801048d1983c46c425d3c95f0cffd@buzz.block.builderlab.xyz>
Co-authored-by: Eva <011987e296fd5006292d2f930b574be47c7801048d1983c46c425d3c95f0cffd@buzz.block.builderlab.xyz>
Search migrated to Postgres FTS (commit f8bbe6efc).
The Typesense container was removed from compose.yml and the Helm chart,
but the cleanup missed two template/config files:
- `deploy/compose/.env.example`: `TYPESENSE_API_KEY` and
`TYPESENSE_PORT` are dead — no typesense service exists in compose.yml
and the relay binary no longer reads `TYPESENSE_API_KEY`. The
`CHANGE_ME_RANDOM_API_KEY` placeholder was never consumed, so removing
it also unbreaks the sed loop in the blog draft (one fewer no-op secret
to generate).
- `benchmarks/harbor-buzz-orchestra/scripts/benchmark.py`: generates a
typesense_api_key in state and writes `TYPESENSE_API_KEY` to the .env
file it creates.
- *Editing this file caused the
https://github.com/block/buzz/blob/main/.github/workflows/benchmark-harbor.yml
linter ci checks to run, which seemingly haven't run before, so I needed
fix the lint issues to pass this.*
---------
Signed-off-by: Kalvin Chau <kalvin@block.xyz>
Co-authored-by: npub1c4alndp82zyt9veaklm5d965quss79vlhk9awv7qu5erwhmf42qqlvc25c <c57bf9b4275088b2b33db7f746975407210f159fbd8bd733c0e532375f69aa80@buzz.block.builderlab.xyz>
