Repository navigation
Description
Source: reading of v0.1.205 (e518363) against v0.1.204 (182 commits, 166 files) on a live Hermes
forward-lane rig. Related issues: #2849, #4168, #2728 (Claude Code), #2372 (Codex), #3451/#1777/#3820/#2193
(dsh), #454 (omp), #1382 (Pi), #2415 (TRAE), #1212, #2219.
The gap
Every other host with a bili plugin lane gets its host-native auto-compaction switched off by the
installer; Hermes is the only one where nothing is written:
- opencode: installer writes the host's own disable key —
src/plugin-install.ts:1441-1448
(compaction.auto = false). - claude: installer injects
CLAUDE_CODE_DISABLE_AUTO_COMPACT=1—src/plugin-install.ts:657-661. - dsh: a dedicated compaction-off overlay ships in-package (
src/claude-web-preset-compat.ts), it latches at
load time (feat: disable dsh native auto-compaction in web profiles at load time (#3354) #3451) and alarms loudly when it cannot latch ([PR] fix: alarm loudly when the dsh compaction-disable overlay cannot latch (#3806) #3820). - omp: launcher disables omp's native auto-compaction (fix: bili omp launcher disables omp's native auto-compaction (#449) #454); pi cancels pi auto-compaction (Pi plugin cancels Pi's auto-compaction for conversations the proxy never sees (breaks pi-claude-bridge) #1382).
- hermes: the setup branch (
src/plugin-install.ts:1814ff) only writes the plugin path and enables
acp_toolsinconfig.yaml. It never touches Hermes' compression settings, so the host keeps
auto-compacting whenever its context fills. In our rig the Hermes-side switch exists and is trivially
settable by hand (compression.auto: false, plus per-providercontext_engine), which makes this a pure
automation gap, not a missing capability.
Consequence: on Hermes lanes two compressors coexist by default. When the host lands a forced compaction
mid-session and the model then switches, the kernel ref space goes stale; we reproduced the hard end of that
on v0.1.202 — after a preflight truncation the model switched, then called compress against a dead range
with an over-wide span, every follow-up call died on range validation, and the session was unusable from the
first message ("[Message truncated]"). We pinned it to host-native compaction (model-switch generation change
leaving a stale fold ledger — same shape as #4038) and had to disable the host compressor by hand to keep the
lane usable. That manual step is now the reason this rig runs the lane at all.
What we ask for
- Setup writes the Hermes disable (parity with opencode/claude/dsh): when the installer wires Hermes in,
writecompression.auto: falseto the Hermes config the installer already reads/writes — including the
profile copy when the session runs under a named profile — and warn at boot if it could not latch
(same pattern as [PR] fix: alarm loudly when the dsh compaction-disable overlay cannot latch (#3806) #3820). Document the Hermes recipe like [PR] docs: tighten the dsh compaction-off recipe (disabled: true + all three presets) #3294 did for dsh. - Emit an explicit host-native-compaction marker. v0.1.205 detects and describes these landings well —
structural finding (src/stream.ts:357), fold-pause on total swap (src/fold-reconcile.ts:2004),
session metadatanativeCompactionAt/nativeCompactionBoundary(src/session.ts:1349-1355), conflict
watcher (src/conflict-watch.ts:59), the web hint "keep exactly one compressor"
(src/web/client.ts:207). None of it reachesacp_statusas a queryable per-session flag. Please add one
(e.g.hostNativeCompression: {at, boundary, hostLane}) so an operator/agent can tell "the host compacted
under bili" from "bili folded", before the next call burns against a dead ref. - Add the Hermes row to the per-client remediation hints (feat: add per-client remediation hints to compression-conflict surfaces (web banner / acp_status) #2219 closed with hints per client; Hermes has
no row in that surface today).
Evidence anchors
Version reading at tag v0.1.205 commit e518363: grep for a native-compaction marker name in src/,
kernel/src/, hermes-plugin/ -> 0 hits; the disable switches above exist for opencode/claude/dsh only.
Rig facts: forward-lane via OpenAI-compatible endpoint, Hermes Python plugin, host auto-compaction enabled by
default; failure transcript and fold-ledger state captured 2026-10-21 on v0.1.202.
Activity
MojitoJoe1813 commentedon Oct 23, 2026
Requested labels (our account has no label permission on this repo):
bug,hermes.ranxianglei commentedon Oct 23, 2026
(ai: vllm-qwen/qwen3.8-27b)
TL;DR (for the reporter): Verified against current master (
v0.1.205, e518363). You're right on both counts — hermes is the only plugin lane whose installer never disables the host's native auto-compaction, and the detection machinery already records native-compaction landings but never exposes them as a queryableacp_statusflag. Two things I need before implementing: (1) your exact hermes config key + hermes version + whether hermes lets a plugin flip compaction off at runtime (you set it by hand, so you know the precise key); (2) the owner's call on where the disable lives (options below). Full analysis in the project's structured form after this.查重: 不重复。hermes 是「setup 阶段关掉宿主原生自动压缩」这个家族里唯一漏掉的车道:opencode(installer 写
compaction.auto=false)、claude(DISABLE_AUTO_COMPACT=1)、dsh(in-package overlay + load-time latch + #3820 报警)、omp(#454)/pi(#1382) launcher 都关了,只有 hermes 的 install 分支(src/plugin-install.ts:1814hermesInstallCore)只拷 Python 插件文件 +hermes plugins enable,完全不碰压缩配置。同家族相关但不重复(客户端/机制各异):claude #2849/#4168、dsh #3294/#3712、pi #1382、omp #454;你复现的死 ref 砖化与 #4038/#2372「宿主原生压缩落在 bili 视野外 → ref space 变 stale」同形状。现有 hermes open issue(#4282 更新不刷 Python 副本、#2714 ephemeral port、#3442 sub-lane)机制都不同。标签:
bughermesp1scope/wiring(缺陷类:默认双压缩器致长会话砖化;p1=会话不可恢复类,虽有手动绕行;主修点在 install/接入车道)。已在镜像打上。我验证了什么(审核三步):
src/plugin-install.ts:1773注释明写 "hermes owns config.yaml, we never parse it" —— 但这只约束 installer(Node 侧);运行时 Python 插件本身早就在读~/.hermes/config.yaml(hermes-plugin/__init__.py:569mitm_domains_from_config、:639_provider_base_url),甚至 import 宿主模块(:768hermes_cli.config_providers)。所以「不解析 config.yaml」不覆盖 runtime 插件 —— 这直接扩大了 fix: keep system prompt byte-stable for prefix cache #1 的方案空间(选项 B 就靠这点)。结构化八问
acp_status里没有可查询标志,operator/agent 赶在下次调用烧到死 ref 前认不出来。nativeCompactionAtsrc/session.ts:1349、nativeCompactionBoundarysrc/session.ts:1353-1384、rebaseNotesrc/stream.ts:432、conflict-watchsrc/conflict-watch.ts:59),但只喂给 conflict 告警和 log,没提升为acp_status可查询的 per-session 状态。.bili-bak/surgical merge 缓解)+ 没有你要的 [PR] fix: alarm loudly when the dsh compaction-disable overlay cannot latch (#3806) #3820 报警。acp_status增量字段{at, boundary, hostLane}(hostLane复用src/acp-status.ts:400已有的 resolved-client);Use kernel renderTags:"text-only" + forbid tag echoing in prompt #3 依 fix: keep system prompt byte-stable for prefix cache #1 定论补 Hermes hint 行(src/conflict-watch.ts:125+per-client 表 +src/web/i18n.ts)。三者同属「宿主原生压缩处理」一个机制族 → 一个 PR 闭环,不拆片。.bili-bak+ remove 还原,否则会吞用户其它 hermes 配置(红线:绝不静默改用户配置)。不动两种压缩模式的 wire 语义。acp_status输出加字段 → 需过 golden snapshot;不改任何既有字段;两种模式都要出(数据两者都有)。compression.auto),不是 bili 配置面;fix: address all code review findings (condense, routing, session-id, dead code) #2 是acp_status输出字段(非配置);B 若走宿主 API 连文件都不写。选项对比(针对 #1)
待决清单(卡在这三点,都不是我能单方拍的)
compression.auto: false吗?还是别的层级?)、你的 hermes 版本、以及 hermes 是否提供插件在运行时改压缩设置的入口。或 owner 让我去翻 hermes 源码确认。写错 key = 静默 no-op,所以这条是 fix: keep system prompt byte-stable for prefix cache #1 的硬前置。nativeCompaction*不止 hermes——Google/Responses/Anthropic/OpenAI prepare 路径 + relay 都会设它(已核实调用点),所以 marker 会是全车道通用观测字段。我建议做成全车道(更值)而非 hermes-only,确认一下?这三点一定,我就开一个 PR 把 #1+#2+#3 一起做完(一个 issue 一个 PR 闭环)。当前不动代码。
MojitoJoe1813 commentedon Oct 23, 2026
Thanks — verified answers for your pending list. Rig Hermes version: Agent v0.22.3 (2026.9.24), local build
5b147ab036d6.(1) Exact key. Hermes has no
compression.autokey. The master switch is top-levelcompression.enabled— documented at https://hermes-agent.nousresearch.com/docs/developer-guide/context-compression-and-caching#configuration ("enabled: true # Enable/disable compression (default: true)"), seeded inhermes_cli/config_defaults.py:592ff, and present in our live~/.hermes/config.yamlline 142 (enabled: true). Verified consumers that honor it:agent/agent_init.py:2020(agent.compression_enabled = cs.enabled) → gates the turn preflight (agent/turn_preflight.py:89,297), idle/turn compaction (agent/turn_context_compaction.py:154,182,240), native-compaction helpers (agent/native_compaction.py:120), andagent/conversation_compression.py:1998. Socompression.enabled: falseis the single invariant-preserving write. Per-route triggers worth naming in the recipe:codex_app_server_auto(native|hermes|off, defaultnative),codex_responses_native(defaultFalse),idle_compact_after_seconds(default0),micro_compact(defaultFalse).(2) Runtime capability — yes, hot reload exists, no restart needed.
_sync_agent_model_with_config()(tui_gateway/model_switch.py:450) runs on the arriving turn (tui_gateway/prompt_turn.py:690) and re-applies live compression config to already-open sessions:tui_gateway/session_compression.py:101setsagent.compression_enabledfrom the raw config value;tests/tui_gateway/test_compression_config_hot_reload.pydocuments exactly this ("adopt live compression config on the next turn" afterhermes config set compression.threshold_tokens …). Supported write path:hermes config set compression.enabled false— boolean coercion via_SCALAR_WORDS(hermes_cli/config.py:3206 _coerce_config_set_value), writing the RAW user config with ruamel comment preservation + section guards + a backups module, so no clobber and no defaults dump. Option B is feasible today with no new Hermes API: the plugin latches at import/load time viaset_config_value, fails open, alarms like #3820 when the read-back is notfalse; open sessions adopt it on their next turn. Option A then stays as a fresh-install convenience write. Our vote: B (A optional), matching the dsh precedent.(3) Marker scope — yes, all lanes. Our plain-http lane and the Codex/Responses/Anthropic server-side paths are the same failure class for clients; a per-session
{at, boundary, hostLane}field is worth more than a Hermes-only field.One caveat worth putting into the remediation text: with
compression.enabled=falsethe preflight only warns on context overflow (agent/turn_preflight.py:212"Uncompressed session guard") and hands an oversized request to the provider un-truncated — on a bili-managed lane that is exactly the desired ownership split (manual/compresskeeps working on the force path,conversation_compression.py:3193,4103), but it means the client-side folder has to stay healthy, because the provider answers400 request exceeds the context window … requests are never truncated. Happy to test a candidate PR against this rig (forward lane, plain-http endpoint, pinned fallback chain).ranxianglei commentedon Oct 23, 2026
(ai: vllm-qwen/qwen3.8-27b)
Thanks — those verified answers unblock most of this, and you caught a real error in my earlier assumption: there is no
compression.autokey; the master switch is top-levelcompression.enabled, and hot reload already exists (session_compression.py:101adopts it on the next turn). That turns #1 from "guess the key" into "target a known, documented switch." Where things stand:已交付:#2 → PR #4459(all-lane marker, as you confirmed)
I shipped ask #2 this turn as PR #4459 — #4459 : a new
HOST-NATIVE COMPACTIONsection inacp_status, shown per-session when a host-native compaction landing is recorded (session.metadata.nativeCompactionBoundary). It names the resolved client/lane, the landing time, and rebase state (NOT yet rebasedvsrebased onto it at <ts>). It is lane-neutral and works in both compression modes (the boundary is set server-side by everyprepare-*/relay path), so hermes/codex/omp/etc. all surface it — matching your "all lanes are the same failure class" point. Additive display-only output (absent when no landing; no wire/config/cache change). Regression test asserts both gate directions; targeted suites green locally (acp-status8/8,codex-official2/2), typecheck clean, full verdict on CI. It references but does not close #4458.#1 + #3:held for owner sign-off + your rig validation
#1 (disable hermes' native auto-compaction) and #3 (hermes hint row) are deliberately not in #4459, for two reasons:
compression.enabled:falseat load"), so it ships together with fix: keep system prompt byte-stable for prefix cache #1.Converged design (my rec = your vote = Option B): the in-repo hermes Python plugin latches at import/load time — set hermes'
compression.enabled=falsethrough the host config path, fail-open (never block hermes boot or model traffic), and emit a #3820-style boot alarm if the read-back isn'tfalse. Open sessions adopt on their next turn via the existing hot reload, no restart. Option A (Node-installer surgical merge-write, opencode-style) stays as a fresh-install convenience/fallback. B preserves the installer's "never parse config.yaml" boundary and delivers the boot alarm you asked for.Your 400 caveat is correct and will go into the remediation/recipe text: with
compression.enabled=false, hermes only warns on overflow and hands the oversized request to the provider un-truncated, so the provider answers400 … requests are never truncated. On a bili lane that's exactly the ownership split we want (bili owns folding; manual/compressstill works on the force path) — but it means bili's client-side folder must stay healthy or the provider 400s. I'll document that tradeoff explicitly in #1's hint row + recipe doc (mirroring #3294).What I need from you (rig validation for #1): against Agent v0.22.3 (build
5b147ab036d6), please confirm the exact load-time entry point the plugin should call to flipcompression.enabled— the preciseset_config_value(...)module/signature — and that a value written before session start is honored on the first turn (not merely adopted later). If you can exercise a candidate PR against your forward-lane / plain-http / pinned-fallback rig, that's precisely the missing validation; I'd rather have you hit the real API than guess. Once you confirm the call site (or the owner says "go read hermes source"), I open the follow-up PR with #1 + #3 together.结构化八问(#4458 现状)
acp_status里没有可查询标志。compression.enabled=false+ fail-open + [PR] fix: alarm loudly when the dsh compaction-disable overlay cannot latch (#3806) #3820 式报警,宿主热载下次 turn 生效);Use kernel renderTags:"text-only" + forbid tag echoing in prompt #3 随 fix: keep system prompt byte-stable for prefix cache #1 补 Hermes hint 行(conflictRemediation + web/client.ts + i18n zh/en)。三者同属「宿主原生压缩处理」一族 → 一个 follow-up PR 闭环。.bili-bak+ remove 还原(绝不静默吞用户配置)。compression.enabled;fix: address all code review findings (condense, routing, session-id, dead code) #2 是acp_status输出字段(非配置)。判定:#2 已进 PR #4459,等 review/CI 即可合;#1+#3 卡在两点——owner 拍板 A/B(我推荐 B) + 你确认 hermes 精确 load-time 调用点(或 owner 让我翻 hermes 源码),两者齐了我就开一个 follow-up PR 把 #1+#3 一起做完。当前不再动代码。
MojitoJoe1813 commentedon Oct 23, 2026
Rig validation for #1, graded so you can see what is measured and what is inference. Rig: Hermes Agent v0.22.3 (2026.9.24), local build
5b147ab036d6. Everything below was read off the installed tree (/root/.hermes/installs/76f6e7e5145b5f39/environments/e4a140c216d7402ab1b11ff54568c00e/workspace), which is the code that actually runs here.1. The exact load-time call site (measured fact)
from hermes_cli.config import set_config_value, live definitiondef set_config_value(key: str, value: str, force: bool = False)athermes_cli/config.py:3406. Call shape for the latch:set_config_value("compression.enabled", "false")._SCALAR_WORDS['false']yields the boolean false via_coerce_config_set_value(hermes_cli/config.py:3198,3206). Noforce=Trueneeded - the key is known, so no wrong-prefix refusal._write_user_configdocstring says "Write only the user's raw config back (never the merged defaults)" (hermes_cli/config.py:3373). Other entries survive.Exceptiondies here:_exit_invalidand_exit_if_key_managedboth callsys.exit(1)(hermes_cli/config.py:3368-3370,:3345ff) - that raisesSystemExit, not anException.is_managed()short-circuits with a print and returns (:230,:3406first lines); on this rig we probed it live and it returnsFalse, so the write path is open here.plugins.load_timeout_seconds, default_LOAD_TIMEOUT_SECS = 10.0(hermes_cli/plugins_loader.py:44-45), clamped at 600 s; after a load timeout every later registration is ignored (PluginContext.mark_load_timed_out). So: latch first, tool registration second, non-blocking, fail-open.register(ctx)inhermes-plugin/__init__.py(installed copy line 1140), whose own file header says nothing may raise out of it. The latch goes inside its existing try-guard, before any tool registration.2. Is a value written before session start honored on the FIRST turn?
agent.compression_enabled = cs.enabled(agent/agent_init.py:2020) - and every auto-compaction gate reads exactly that attribute:agent/turn_preflight.py:89,297,agent/turn_context_compaction.py:154,240,agent/native_compaction.py:120. Hot reload for already-open sessions exists as described (tui_gateway/prompt_turn.py:691->tui_gateway/session_compression.py:174->_apply_live_compression_config, assignment at:101), short-circuiting only when the config signature is unchanged.tests/tui_gateway/test_compression_config_hot_reload.py); note that test ships in the checkout, not inside the installed tree.model_tools.py:162and at gateway startup (gateway/config.py:577,gateway/run_startup.py:1078), so in practice a load-time latch is armed before any session agent exists; the TUI agent factory calls discovery only inside a toolset-pin branch (tui_gateway/agent_factory.py:354). We rate "latch precedes first agent construction on this rig" at 85 %, and the [PR] fix: alarm loudly when the dsh compaction-disable overlay cannot latch (#3806) #3820-style read-back alarm you planned is exactly the right cover for the residual risk.3. PR #4459: reviewed against the v0.1.205 tree - ship it
markNativeCompactionBoundary()(src/session.ts:1353) is called fromsrc/server/prepare-openai.ts:178(plain-http forward - our shape),src/server/relay.ts:1738(attach relay),src/server/prepare-anthropic.ts:286; the rebase transition runs atprepare-openai.ts:179,prepare-anthropic.ts:287,prepare-responses.ts:70. So the "NOT yet rebased" vs "rebased onto it at" states are produced server-side per request path, not synthesized in the display layer.x-bili-plugin: hermes(hermes-plugin/__init__.py:44, header merge at:1070), bili records it assession.metadata.pluginAgent(src/server/handle.ts:836,1496), andconflictClientOf()returns it (src/conflict-watch.ts:111-119). On this rig your line reads "the client (hermes) compacted this conversation natively ...". Rendering we have not seen live, so: chain measured, display expected (95 %).extra.push(...)plus a test with a negative gate asserting no marker when no landing was recorded). No wire/cache/fold semantics touched. Good as-is.ProxyToolResultalready carries machine-readable fields (outcome,reason,src/proxy-tool-result.ts:37-46) while this marker is prose. A matching optional field on the acp_status result, e.g.nativeCompaction?: { at: number; pendingRebase: boolean; rebasedAt?: number; lane: string }, would let the panel/relay read the boundary without parsing text. Happy to test that variant if you add it.4. Exercise a candidate PR on our rig - we cannot do that today, and here is why, plainly
We took billion-context out of operation on 22 Oct. The reason was not one proven bili defect: two chat symptoms kept recurring (a
compresscall that failed with a JSON-shaped refusal, and old prompt content resurfacing later in the chat), and the measured root causes were Hermes-side tool deferral/pinning plus host-native compaction rewriting the history under bili - i.e. exactly the double-compaction class this issue describes. Taking the lane out stopped the symptom for everyday chats.So: your #1/#3 confirmation stands on source measurement (sections 1-2), and a candidate build can be reviewed the same way we reviewed #4459 - line by line against a tag tree.
5. One correction to my previous comment
I cited
agent/conversation_compression.py:3193,4103for "manual/compresskeeps working on the force path". Those two lines prove the force path bypasses the thrash latch and the cooldown. The absence of the master gate on the manual path is proven by two other places:gateway/slash_commands_session.py:685-760andagent/compression_facade.py:227never consultcompression_enabled. Behaviour as such: 90 % (source-level, not run here). Your recipe text should cite those, so nobody reads it as "force flips the master switch off-guard".ranxianglei commentedon Oct 23, 2026
(ai: vllm-qwen/qwen3.8-27b)
Outstanding validation — the graded measured/inference split, the
SystemExitgotcha, and the load-timeout constraint are exactly what I needed. Two things first:Correction accepted (§5): you're right —
conversation_compression.py:3193,4103prove the force path bypasses the thrash latch/cooldown, but the absence of the master gate on the manual path is proven bygateway/slash_commands_session.py:685-760andagent/compression_facade.py:227never consultingcompression_enabled. The #3 recipe/hint text will cite those two, so nobody reads it as "force flips the master switch off-guard."PR #4459 — keeping it as-is (your "good as-is" stands)
Thanks for the line-by-line review against v0.1.205 — the server-side boundary production and the
(hermes)lane resolution matching your rig is strong confirmation. On the optional machine-readable field: I'm deliberately not adding it to this PR.acp_statusreturns a single prosetextblob (toolOk(...)atsrc/acp-status.ts:508), and every diagnostic in it is a stable-prefix line (DEAD REFS —,COMPRESS CIRCUIT BREAKER: ARMED —, …). MyHOST-NATIVE COMPACTION — <lane> … at <ts>line follows that exact idiom and is grep-parseable the same way. A true structured field would mean extending the sharedProxyToolResultenvelope (src/proxy-tool-result.ts:37) with an acp_status-only member plus teaching the forwarding layer to carry it — a cross-cutting contract change that's not warranted for an "optional, not a blocker" nicety inside this additive PR. If you want envelope-level parsing, that's a clean small follow-up; happy to take it on its own.#4459otherwise awaits owner merge (merge is human-only) after CI.#1 — fully specced from your measurements; needs the owner's go
Everything for Option B is now pinned down. The implementation I'd build (and re-verify against the real hermes tree while doing it — I won't ship an external account's line numbers blind):
register(ctx)inhermes-plugin/__init__.py, within its existing try-guard, before any tool registration (latch first, registration second — keeps it inside theplugins.load_timeout_secondsdefault 10 s budget).from hermes_cli.config import set_config_value→set_config_value("compression.enabled", "false")(host-side_SCALAR_WORDS['false']coercion; noforce=True; raw-config-only write, other entries survive)._exit_invalid/_exit_if_key_managedcallsys.exit(1)→ raisesSystemExit, notException. The fail-open guard must catch(Exception, SystemExit)(orBaseException), log + alarm, and continue — catching onlyExceptionwould kill the load.false, emit a startup alarm. This covers the 85 % residual you rated for "latch precedes first agent construction."My recommendation = B (= your vote, = dsh #3820 precedent): preserves the installer's "never parse config.yaml" boundary, delivers the boot alarm you asked for, self-heals every start. Option A (Node-installer surgical merge-write, opencode-style) stays as a fresh-install convenience/fallback. #3 (Hermes hint row across
conflictRemediation+web/client.ts+ i18n zh/en) ships in the same follow-up PR. One issue → one follow-up PR, closed loop.Urgency noted: you took the lane offline on 22 Oct because of exactly this double-compaction class. Understood — the moment the owner confirms B, I'll have the follow-up PR up the same day.
结构化八问(#1 决策)
SystemExit+ 须在 load timeout 内非阻塞;好处=[PR] fix: alarm loudly when the dsh compaction-disable overlay cannot latch (#3806) #3820 报警 + 每次启动自愈 + 保住 installer 边界 + open sessions 下 turn 生效(95%)。compression.enabled=false,catch(Exception,SystemExit),fail-open,读回报警);Use kernel renderTags:"text-only" + forbid tag echoing in prompt #3 随 PR 补 Hermes hint 行,文案引用你更正后的手动路径位置。.bili-bak+ remove 还原(绝不静默吞用户配置);B=插件 load 新增行为,严格 fail-open,读回失败/超时都不阻断启动或 traffic,两种压缩模式语义不变;Use kernel renderTags:"text-only" + forbid tag echoing in prompt #3=纯 hint 文案,三处同步 zh/en 一致。compression.enabled。判定:#2 已进 PR #4459(等 owner 合 + CI);#1+#3 的全部实现细节已按你的实测钉死,只差 owner 拍板 A/B(我推荐 B)——你一句「B,干」我今天就开 follow-up PR。当前不再动代码。
MojitoJoe1813 commentedon Oct 23, 2026
B, go. Option B confirmed - please build it exactly as you pinned it to our measurements.
(Exception, SystemExit)on the write path (_exit_invalid/_exit_if_key_managed->sys.exit(1)), log + alarm, continue. Catching onlyExceptionkills the load.gateway/slash_commands_session.py:685-760,agent/compression_facade.py:227) and keep the oversized-request caveat in the hint text ("with host compaction off, oversized requests reach the provider un-truncated").Structured field: agreed, it does not belong in #4459 - a shared-envelope change for a display nicety is the wrong trade. Open your own follow-up for it whenever you want; we will review and test that variant on its own merits.
From this rig today: when the follow-up ships, I review it line by line against a tag tree (same method as #4459) and report measured facts separately from inference, with confidence stated. What I cannot do from this rig right now is a live run - our lane is offline - so treat any "we exercised it" claim as pending until the lane comes back under your #1/#3 build.
Ship it.
ranxianglei commentedon Oct 23, 2026
(ai: vllm-qwen/qwen3.8-27b)
Thanks — and thanks for agreeing the structured field belongs in its own follow-up, and for the line-by-line / measured-vs-inference review commitment. Noted that live runs are pending until the lane is back under the #1/#3 build, so I'll treat any "we exercised it" claim as pending too.
One process point before I cut the branch: #1 is a behavior change — bili would now mutate the hermes host config (
compression.enabled=false) at plugin load, plus add a load-time latch inside the plugin. Per the project rule (and the gate I set for #1), that A/B call is the owner's to make, not a contributor's. Your confirmation is strong corroboration and I independently recommend B, but I'm not going to start a behavior-change implementation on an external account's "go" alone — the deciding vote has to come from the owner.@ranxianglei — one word from you ("B, go") and I build the follow-up PR today, exactly as pinned last turn:
register(ctx)before tool registration, inside the 10 s load budget, non-blocking, fail-open;set_config_value("compression.enabled","false"), catch(Exception, SystemExit)→ log + alarm + continue (catching onlyExceptionkills the load);conflictRemediation+web/client.ts+ i18n zh/en) citing the agreed manual-path lines (slash_commands_session.py:685-760,compression_facade.py:227) + the oversized-request caveat ("with host compaction off, oversized requests reach the provider un-truncated").Decision-relevant recap for your call:
compression.enabled.Standing by for your sign-off.
feat(acp-status): surface host-native compaction landings as an expli…