fix(buzz-agent): scope handoff cap per turn, not per session lifetime - #4805
Merged
Conversation
BUZZ_AGENT_MAX_HANDOFFS previously compared against the session-cumulative `handoff_count` (persisted across prompts). After 10 successful compactions a long-lived session hit the cap permanently: `maybe_handoff()` returned Skipped on every subsequent prompt, the 16 MiB byte-truncation fallback never bound before a 1 M-token provider wall, and the session wedged on the first 400 with no recovery path. Fix: replace the session-level cap check with a local `handoff_attempts` counter constructed at the start of `run()` and passed into `maybe_handoff()`. The counter resets on every `session/prompt` turn so BUZZ_AGENT_MAX_HANDOFFS caps compaction loops within a single turn while allowing unbounded compactions across a session's lifetime. The session-cumulative `handoff_count` is retained for log context only and is not reset. Steer-driven rounds share the same per-turn budget automatically (steers inject into the running `run()` loop, not a new call). Additional changes: - Increment `handoff_attempts` before `summarize()` so failed, empty, and cancelled summarise attempts each consume one slot — the cap cannot be bypassed by a repeatedly-failing summarizer. - Upgrade cap-forced Skipped from INFO to WARN; include session_id, attempt count, projected tokens, and token threshold in structured fields so the cap→wall pairing is attributable per session. - Document `max_handoffs` (config.rs) as a per-turn bound. Three new regression tests: - `handoff_cap_resets_per_turn_not_per_session`: two turns each compact with cap=1, proving the lifetime kill switch is gone. - `handoff_cap_binds_within_a_single_turn`: multi-round turn with cap=1; round 1 preflight emits the cap WARN after round 0 consumed the slot. - `failed_summarize_burns_handoff_attempt_budget`: a failing summarize in round 0 still increments the counter; round 1 preflight sees cap hit, proving pre-summarize accounting. Co-authored-by: Will Pfleger <pfleger.will@gmail.com> Signed-off-by: Will Pfleger <pfleger.will@gmail.com>
…sion Add required structured tracing fields to the handoff cap WARN: - reason = "preflight" (disambiguates from future reactive caps) - handoff_attempts = N (attempt count as a structured field) - max_handoffs = N (cap limit as a structured field) - Drop duplicate prose from message; keep session_id, projected_tokens, threshold_tokens unchanged. Extend handoff_cap_binds_within_a_single_turn to inject a mid-turn steer and assert it is accepted in the live run. The test then proves the next round's threshold crossing is still cap-blocked with no second summarize call — verifying handoff_attempts is not reset by the steer path. Pin the new structured fields in the cap regression so a field rename breaks the test. Co-authored-by: Will Pfleger <pfleger.will@gmail.com> Signed-off-by: Will Pfleger <pfleger.will@gmail.com>
wpfleger96
pushed a commit
that referenced
this pull request
Aug 5, 2026
…-overflow-recovery * origin/main: fix(buzz-agent): scope handoff cap per turn, not per session lifetime (#4805) Fix mobile message timeline bounce (#4862) Polish mobile bottom sheets and profile cards (#4911) Fix media attachment actions (#4849) fix(desktop): remove join API token control (#4897) fix(desktop): allow shared agent mentions (#4913) Polish mobile top navigation (#4778) fix(release): tag immutable desktop candidates (#4811) fix(channels): restrict private-channel invitations (#4612) fix(acp): reject unattended permission requests (#4609) fix(workflow): bind trigger author to the signed event (#4607) fix(git): revoke access for banned relay members (#4608) fix(agent): recover from unsupported image input instead of poisoning the turn (#4896) Define private managed agent wire protocol (#4593) fix(mobile): serialize channel sections sync (#3165) fix(desktop): make missing-command error actionable for released builds (#4802) chore(release): release Buzz Desktop version 0.5.5 (#4809) feat: paste composer text without formatting (#4801) Revert "chore(release): release Buzz Desktop version 0.5.5" (#4808) chore(release): release Buzz Desktop version 0.5.5 (#4800) Co-authored-by: Will Pfleger <pfleger.will@gmail.com> Signed-off-by: Will Pfleger <pfleger.will@gmail.com> # Conflicts: # crates/buzz-agent/src/agent.rs # crates/buzz-agent/src/handoff.rs # crates/buzz-agent/src/types.rs # crates/buzz-agent/tests/regressions.rs
atishpatel
added a commit
that referenced
this pull request
Aug 5, 2026
…usage * origin/main: fix(buzz-agent): scope handoff cap per turn, not per session lifetime (#4805) Fix mobile message timeline bounce (#4862) Polish mobile bottom sheets and profile cards (#4911) Fix media attachment actions (#4849) fix(desktop): remove join API token control (#4897) fix(desktop): allow shared agent mentions (#4913) Polish mobile top navigation (#4778) fix(release): tag immutable desktop candidates (#4811) fix(channels): restrict private-channel invitations (#4612) fix(acp): reject unattended permission requests (#4609) fix(workflow): bind trigger author to the signed event (#4607) fix(git): revoke access for banned relay members (#4608) Co-authored-by: Atish Patel <atish@squareup.com> Signed-off-by: Atish Patel <atish@squareup.com>
wpfleger96
added a commit
that referenced
this pull request
Aug 5, 2026
…#4946) Provider `context_length_exceeded` 400s permanently wedged agent sessions: the turn errored, the oversized history persisted in the in-memory session, and the usage baseline stayed frozen at the last successful sub-threshold reading (failed requests report no usage), so the preflight handoff gate never fired again — every later prompt failed identically until an agent restart. The byte-truncation fallback never intervened because it is a request-body limiter (`estimated_bytes`), not a context-window defence; at context-window scale it is a measured no-op. This adds the reactive recovery path: - **Typed classification.** `AgentError::LlmContextExceeded` is classified at both non-success provider terminals — the shared `post()` (Anthropic, OpenAI, Databricks) and `openrouter_post()` — on `status == 400` plus a context-window body match, so ordinary 400s stay terminal. - **Forced handoff.** A context-400 forces a summarize-handoff that bypasses `should_handoff()` and `BUZZ_AGENT_MAX_HANDOFFS`, bounded by its own per-turn budget (`MAX_CONTEXT_RECOVERIES_PER_RUN = 3`). - **Shrink ladder.** The summarize prompt budget halves from the observed rejected history size — not from `max_context_tokens`, the number the provider just contradicted — rung to rung, with a 4096-byte floor. A summarize call rejected for the same reason takes the next rung instead of re-sticking. At the floor (overflow dominated by unshrinkable frame: system prompt, tool schemas, live prompt) recovery is refused and the provider error surfaces clearly instead of self-healing. - **Baseline reset.** The stale usage baseline is cleared when a request fails, so the preflight gate cannot stay frozen sub-threshold on retries. Named behavior changes: 1. **Anthropic and OpenRouter errors now carry the `(model)` stamp.** Provider arms return their `Result` into the central error mapper instead of early-returning past it, making the code match its documented single-convergence contract at that mapper. 2. **`max_rounds` now counts completions the loop acts on.** A request rejected with a context-400 that is then successfully recovered refunds its round before the retry, paired 1:1 with a consumed recovery rung, so the round cap is neither weakened nor able to drop a recovered turn unanswered. Related: #4805 — the complementary proactive fix (per-session handoff-cap kill switch that let sessions grow to the provider wall). #4805 prevents reaching the wall; this PR recovers at it. --------- Co-authored-by: npub17jjz49l9jjmhhk7cac63j8yt9z555n9cw8vk7v5jz4vzw4ppld5qgj57cc <f4a42a97e594b77bdbd8ee35191c8b28a94a4cb871d96f32921558275421fb68@buzz.block.builderlab.xyz> Co-authored-by: Duncan <dcfd242e557282d7a1e2cf2e6877522682f1e5c6156dc92ca7d90eaedd3b0f95@buzz.block.builderlab.xyz>
wpfleger96
pushed a commit
that referenced
this pull request
Aug 5, 2026
…arer-auth * origin/main: (65 commits) fix(desktop): route macos notification clicks (#4799) feat(mobile): sync themes per community (#3767) feat(desktop): sync themes per community (#3653) feat(desktop): cap OpenClaw agent parallelism at 5 (#4019) fix(buzz-agent): scope handoff cap per turn, not per session lifetime (#4805) Fix mobile message timeline bounce (#4862) Polish mobile bottom sheets and profile cards (#4911) Fix media attachment actions (#4849) fix(desktop): remove join API token control (#4897) fix(desktop): allow shared agent mentions (#4913) Polish mobile top navigation (#4778) fix(release): tag immutable desktop candidates (#4811) fix(channels): restrict private-channel invitations (#4612) fix(acp): reject unattended permission requests (#4609) fix(workflow): bind trigger author to the signed event (#4607) fix(git): revoke access for banned relay members (#4608) fix(agent): recover from unsupported image input instead of poisoning the turn (#4896) Define private managed agent wire protocol (#4593) fix(mobile): serialize channel sections sync (#3165) fix(desktop): make missing-command error actionable for released builds (#4802) ... Co-authored-by: Will Pfleger <pfleger.will@gmail.com> Signed-off-by: Will Pfleger <pfleger.will@gmail.com> # Conflicts: # CHANGELOG.md
atishpatel
added a commit
that referenced
this pull request
Aug 5, 2026
…usage * origin/main: fix(buzz-agent): scope handoff cap per turn, not per session lifetime (#4805) Fix mobile message timeline bounce (#4862) Polish mobile bottom sheets and profile cards (#4911) Fix media attachment actions (#4849) fix(desktop): remove join API token control (#4897) fix(desktop): allow shared agent mentions (#4913) Polish mobile top navigation (#4778) fix(release): tag immutable desktop candidates (#4811) fix(channels): restrict private-channel invitations (#4612) fix(acp): reject unattended permission requests (#4609) fix(workflow): bind trigger author to the signed event (#4607) fix(git): revoke access for banned relay members (#4608) Co-authored-by: Atish Patel <atish@squareup.com> Signed-off-by: Atish Patel <atish@squareup.com>
wpfleger96
pushed a commit
that referenced
this pull request
Aug 5, 2026
…ed-agent-store-merge * origin/main: (24 commits) fix(reactions): support max-length custom emoji (#3833) feat(desktop): allow leaving your final community (#3621) fix(buzz-agent): recover from context-window 400s instead of sticking (#4946) docs(persona-pack): fix stale desktop import instructions (#4500) fix(desktop): route macos notification clicks (#4799) feat(mobile): sync themes per community (#3767) feat(desktop): sync themes per community (#3653) feat(desktop): cap OpenClaw agent parallelism at 5 (#4019) fix(buzz-agent): scope handoff cap per turn, not per session lifetime (#4805) Fix mobile message timeline bounce (#4862) Polish mobile bottom sheets and profile cards (#4911) Fix media attachment actions (#4849) fix(desktop): remove join API token control (#4897) fix(desktop): allow shared agent mentions (#4913) Polish mobile top navigation (#4778) fix(release): tag immutable desktop candidates (#4811) fix(channels): restrict private-channel invitations (#4612) fix(acp): reject unattended permission requests (#4609) fix(workflow): bind trigger author to the signed event (#4607) fix(git): revoke access for banned relay members (#4608) ... Co-authored-by: Will Pfleger <pfleger.will@gmail.com> Signed-off-by: Will Pfleger <pfleger.will@gmail.com> # Conflicts: # desktop/src-tauri/src/managed_agents/runtime.rs
elifoster-block
added a commit
that referenced
this pull request
Aug 5, 2026
…p-csp * origin/main: (66 commits) fix(reactions): support max-length custom emoji (#3833) feat(desktop): allow leaving your final community (#3621) fix(buzz-agent): recover from context-window 400s instead of sticking (#4946) docs(persona-pack): fix stale desktop import instructions (#4500) fix(desktop): route macos notification clicks (#4799) feat(mobile): sync themes per community (#3767) feat(desktop): sync themes per community (#3653) feat(desktop): cap OpenClaw agent parallelism at 5 (#4019) fix(buzz-agent): scope handoff cap per turn, not per session lifetime (#4805) Fix mobile message timeline bounce (#4862) Polish mobile bottom sheets and profile cards (#4911) Fix media attachment actions (#4849) fix(desktop): remove join API token control (#4897) fix(desktop): allow shared agent mentions (#4913) Polish mobile top navigation (#4778) fix(release): tag immutable desktop candidates (#4811) fix(channels): restrict private-channel invitations (#4612) fix(acp): reject unattended permission requests (#4609) fix(workflow): bind trigger author to the signed event (#4607) fix(git): revoke access for banned relay members (#4608) ...
tellaho
added a commit
that referenced
this pull request
Aug 5, 2026
Co-authored-by: Taylor Ho <taylorkmho@gmail.com> Signed-off-by: Taylor Ho <taylorkmho@gmail.com> * origin/main: fix(reactions): support max-length custom emoji (#3833) feat(desktop): allow leaving your final community (#3621) fix(buzz-agent): recover from context-window 400s instead of sticking (#4946) docs(persona-pack): fix stale desktop import instructions (#4500) fix(desktop): route macos notification clicks (#4799) feat(mobile): sync themes per community (#3767) feat(desktop): sync themes per community (#3653) feat(desktop): cap OpenClaw agent parallelism at 5 (#4019) fix(buzz-agent): scope handoff cap per turn, not per session lifetime (#4805) Fix mobile message timeline bounce (#4862) Polish mobile bottom sheets and profile cards (#4911) Fix media attachment actions (#4849) fix(desktop): remove join API token control (#4897) fix(desktop): allow shared agent mentions (#4913) Signed-off-by: Carl <acda9e433d19dcd0e6b6840f7f4b98f3a56f1fab98049d444c087019e6d36560@buzz.block.builderlab.xyz>
wpfleger96
added a commit
that referenced
this pull request
Aug 5, 2026
Co-authored-by: Will Pfleger <pfleger.will@gmail.com> Signed-off-by: Will Pfleger <pfleger.will@gmail.com> * origin/main: fix(reactions): support max-length custom emoji (#3833) feat(desktop): allow leaving your final community (#3621) fix(buzz-agent): recover from context-window 400s instead of sticking (#4946) docs(persona-pack): fix stale desktop import instructions (#4500) fix(desktop): route macos notification clicks (#4799) feat(mobile): sync themes per community (#3767) feat(desktop): sync themes per community (#3653) feat(desktop): cap OpenClaw agent parallelism at 5 (#4019) fix(buzz-agent): scope handoff cap per turn, not per session lifetime (#4805) Fix mobile message timeline bounce (#4862) Polish mobile bottom sheets and profile cards (#4911) Fix media attachment actions (#4849) fix(desktop): remove join API token control (#4897) Signed-off-by: Duncan <dcfd242e557282d7a1e2cf2e6877522682f1e5c6156dc92ca7d90eaedd3b0f95@buzz.block.builderlab.xyz> Co-authored-by: Will Pfleger <pfleger.will@gmail.com> Signed-off-by: Will Pfleger <pfleger.will@gmail.com>
tellaho
added a commit
that referenced
this pull request
Aug 5, 2026
…-setting * origin/main: fix(reactions): support max-length custom emoji (#3833) feat(desktop): allow leaving your final community (#3621) fix(buzz-agent): recover from context-window 400s instead of sticking (#4946) docs(persona-pack): fix stale desktop import instructions (#4500) fix(desktop): route macos notification clicks (#4799) feat(mobile): sync themes per community (#3767) feat(desktop): sync themes per community (#3653) feat(desktop): cap OpenClaw agent parallelism at 5 (#4019) fix(buzz-agent): scope handoff cap per turn, not per session lifetime (#4805) Fix mobile message timeline bounce (#4862) Polish mobile bottom sheets and profile cards (#4911) Fix media attachment actions (#4849) fix(desktop): remove join API token control (#4897) Co-authored-by: Taylor Ho <taylorkmho@gmail.com> Signed-off-by: Taylor Ho <taylorkmho@gmail.com>
sandro-sq
added a commit
that referenced
this pull request
Aug 5, 2026
* origin/main: (32 commits) fix(mobile): merge relay recounts with locally seen thread replies (#4633) fix(desktop): enable message editing in Inbox (#2198) relay: fuzz WebSocket 1012 restart-close timing on graceful drain (BUZZ_DRAIN_JITTER_MS) (#4542) fix(desktop): outline the selected community (#4969) fix(desktop): clamp thread panel to channel surface (#4965) style(messages): increase username contrast (#4948) fix(desktop): rename generic attachment action from 'Attach image' to 'Attach file' (#2381) (#4304) fix(reactions): support max-length custom emoji (#3833) feat(desktop): allow leaving your final community (#3621) fix(buzz-agent): recover from context-window 400s instead of sticking (#4946) docs(persona-pack): fix stale desktop import instructions (#4500) fix(desktop): route macos notification clicks (#4799) feat(mobile): sync themes per community (#3767) feat(desktop): sync themes per community (#3653) feat(desktop): cap OpenClaw agent parallelism at 5 (#4019) fix(buzz-agent): scope handoff cap per turn, not per session lifetime (#4805) Fix mobile message timeline bounce (#4862) Polish mobile bottom sheets and profile cards (#4911) Fix media attachment actions (#4849) fix(desktop): remove join API token control (#4897) ... Signed-off-by: Alessandro Joabar <sandro@squareup.com>
wpfleger96
pushed a commit
that referenced
this pull request
Aug 5, 2026
…-agents-nav * origin/main: fix(mobile): merge relay recounts with locally seen thread replies (#4633) fix(desktop): enable message editing in Inbox (#2198) relay: fuzz WebSocket 1012 restart-close timing on graceful drain (BUZZ_DRAIN_JITTER_MS) (#4542) fix(desktop): outline the selected community (#4969) fix(desktop): clamp thread panel to channel surface (#4965) style(messages): increase username contrast (#4948) fix(desktop): rename generic attachment action from 'Attach image' to 'Attach file' (#2381) (#4304) fix(reactions): support max-length custom emoji (#3833) feat(desktop): allow leaving your final community (#3621) fix(buzz-agent): recover from context-window 400s instead of sticking (#4946) docs(persona-pack): fix stale desktop import instructions (#4500) fix(desktop): route macos notification clicks (#4799) feat(mobile): sync themes per community (#3767) feat(desktop): sync themes per community (#3653) feat(desktop): cap OpenClaw agent parallelism at 5 (#4019) fix(buzz-agent): scope handoff cap per turn, not per session lifetime (#4805) Fix mobile message timeline bounce (#4862) Polish mobile bottom sheets and profile cards (#4911) Fix media attachment actions (#4849) fix(desktop): remove join API token control (#4897) Signed-off-by: Duncan <dcfd242e557282d7a1e2cf2e6877522682f1e5c6156dc92ca7d90eaedd3b0f95@buzz.block.builderlab.xyz>
brow
pushed a commit
that referenced
this pull request
Aug 6, 2026
…rebase-wt * origin/main: (29 commits) fix(mobile): merge relay recounts with locally seen thread replies (#4633) fix(desktop): enable message editing in Inbox (#2198) relay: fuzz WebSocket 1012 restart-close timing on graceful drain (BUZZ_DRAIN_JITTER_MS) (#4542) fix(desktop): outline the selected community (#4969) fix(desktop): clamp thread panel to channel surface (#4965) style(messages): increase username contrast (#4948) fix(desktop): rename generic attachment action from 'Attach image' to 'Attach file' (#2381) (#4304) fix(reactions): support max-length custom emoji (#3833) feat(desktop): allow leaving your final community (#3621) fix(buzz-agent): recover from context-window 400s instead of sticking (#4946) docs(persona-pack): fix stale desktop import instructions (#4500) fix(desktop): route macos notification clicks (#4799) feat(mobile): sync themes per community (#3767) feat(desktop): sync themes per community (#3653) feat(desktop): cap OpenClaw agent parallelism at 5 (#4019) fix(buzz-agent): scope handoff cap per turn, not per session lifetime (#4805) Fix mobile message timeline bounce (#4862) Polish mobile bottom sheets and profile cards (#4911) Fix media attachment actions (#4849) fix(desktop): remove join API token control (#4897) ... Co-authored-by: Tom Brow <tomb@block.xyz> Signed-off-by: Tom Brow <tomb@block.xyz> # Conflicts: # desktop/src-tauri/src/commands/agents.rs # desktop/src-tauri/src/commands/agents_deploy.rs # desktop/src-tauri/src/managed_agents/env_vars.rs
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.
BUZZ_AGENT_MAX_HANDOFFScompared against the session-cumulativehandoff_count(persisted across prompts). After N handoffs a long-lived session hit the cap permanently:maybe_handoff()returnedSkippedon every subsequent prompt, the 16 MiB byte-truncation fallback never bound before a 1M-token provider wall, and the session wedged on the first 400 with no recovery path. Thufir's session log shows 8 days of cap-forced truncation before the firstcontext_length_exceeded400.The fix replaces the session-level cap comparison with a local
handoff_attemptscounter constructed at the start ofrun()and passed intomaybe_handoff(). The counter resets on everysession/promptturn soBUZZ_AGENT_MAX_HANDOFFScaps compaction loops within a single turn while allowing unbounded compactions across a session's lifetime. The session-cumulativehandoff_countis retained for log context only and is not reset. Steer-driven rounds share the per-turn budget automatically since steers inject into the runningrun()loop, not a new call.handoff_attemptsincrement to beforesummarize()so failed, empty, and cancelled summarize calls each consume one budget slot — the cap cannot be bypassed by a repeatedly-failing summarizerSkippedfromINFOtoWARN; add structured fields forsession_id, attempt count, projected tokens, and threshold so the cap→wall pairing is attributable per sessionmax_handoffsinconfig.rsas a per-session/prompt-turn boundNote: this is the proactive half of the context-window fix. The reactive
context_length_exceeded400 recovery path is owned by Sami's branch (buzz-ctxfix-sami, Tyler's crew); this PR is intended to land after that one.