Teknium 543d305bbb feat(moa): add reference_max_tokens to cap advisor output and cut turn latency (#56756) ... MoA per-turn latency is dominated by advisor GENERATION: turn wall time correlates ~0.88 with output tokens and ~-0.03 with input tokens (measured over 52 turns). Each turn waits for the slowest advisor to finish writing, and advisors were uncapped — writing multi-thousand-token essays the aggregator only needs the gist of. Add an opt-in per-preset reference_max_tokens knob (mirrors reference_temperature) that caps ADVISOR output only; the acting aggregator is never capped. Default None = uncapped, so existing presets are byte-for-byte unchanged (no regression). Wired through both MoA execution paths (MoAChatCompletions.create and aggregate_moa_context). E2E: same task, closed preset uncapped vs reference_max_tokens=600 -> 59s to 33s (~44% faster), final answer identical/correct. - hermes_cli/moa_config.py: _coerce_int_or_none helper + reference_max_tokens in _normalize_preset/_default_preset/flattened view - agent/moa_loop.py: read preset.reference_max_tokens, pass to reference fan-out - agent/conversation_loop.py: pass reference_max_tokens on the per-turn path - tests + docs		2026-07-02 00:16:35 -07:00
..
lsp	fix(tui): prevent killpg suicide during MCP shutdown	2026-07-01 04:54:46 -07:00
pet	fix(pet): snap kitty frames to whole cells	2026-06-30 15:41:44 -05:00
secret_sources
transports	fix(security): strip dynamic Hermes secrets from all subprocess spawn env	2026-07-01 14:37:22 +05:30
__init__.py
account_usage.py
agent_init.py	fix(agent): honor custom CA certs on aux client + harden TLS resolution	2026-07-02 04:51:56 +05:30
agent_runtime_helpers.py	fix(agent): honor custom CA certs on aux client + harden TLS resolution	2026-07-02 04:51:56 +05:30
anthropic_adapter.py	fix(anthropic): use claude-code/ UA prefix for OAuth to avoid 404 (#48534)	2026-07-01 15:42:15 +05:30
async_utils.py
auxiliary_client.py	fix(moa): lift hidden Anthropic aux output cap	2026-07-02 06:31:18 +05:30
azure_identity_adapter.py
background_review.py	fix(review): isolate the background-review fork from the canonical session	2026-07-01 16:21:39 +05:30
bedrock_adapter.py
billing_view.py	feat(billing): /billing terminal billing — interactive TUI + CLI client (#45449)	2026-06-19 01:53:32 +05:30
browser_provider.py
browser_registry.py
chat_completion_helpers.py	fix(streaming): handle completed responses with empty/None choices (#55933) (#56713)	2026-07-02 06:36:20 +05:30
codex_responses_adapter.py
codex_runtime.py	refactor(approval): extract is_approval_bypass_active(); use frozen-env bypass in codex routing	2026-07-01 22:58:37 +05:30
coding_context.py	feat(agent): add configurable coding_instructions	2026-06-30 00:59:59 -05:00
context_breakdown.py	feat(desktop): add context usage breakdown popover	2026-06-29 09:18:10 -04:00
context_compressor.py	fix(compaction): detect and strip merge-into-tail summaries past the delimiter	2026-07-01 18:23:01 +05:30
context_engine.py	fix(context): clamp -1 post-compression sentinel in sibling status paths	2026-07-01 13:36:50 +05:30
context_references.py	fix(security): anchor @file context refs to canonical read deny-list	2026-07-01 02:43:49 -07:00
conversation_compression.py	fix(agent): make compression lock-lease refresher tolerate transient DB blips	2026-06-30 13:36:29 +05:30
conversation_loop.py	feat(moa): add reference_max_tokens to cap advisor output and cut turn latency (#56756)	2026-07-02 00:16:35 -07:00
copilot_acp_client.py	fix(agent): stream copilot ACP chat completions	2026-06-28 22:52:51 -07:00
credential_persistence.py
credential_pool.py	fix(auth): serialize Codex OAuth pool refresh under the auth-store lock (#56233)	2026-07-01 02:45:07 -07:00
credential_sources.py
credits_tracker.py
curator.py	fix(curator): never archive cron-referenced skills + floor use=0 pruning (#54443)	2026-06-28 15:10:21 -07:00
curator_backup.py
display.py	feat(display): friendly human-phrased tool labels for built-in tools (#55166)	2026-06-29 20:31:17 -07:00
error_classifier.py	fix(error-classifier): route 5xx context-overflow into compression	2026-07-01 16:14:16 +05:30
errors.py
file_safety.py	fix(file): block credential paths from search results	2026-07-01 01:02:35 -07:00
gemini_native_adapter.py	Merge consecutive same-role contents for native Gemini	2026-06-30 11:51:22 -07:00
gemini_schema.py
i18n.py
image_gen_provider.py	feat(image-gen): add image-to-image / editing to image_generate (#48705)	2026-06-18 22:13:07 -07:00
image_gen_registry.py
image_routing.py	fix(vision): detect Ollama vision models via /api/show (#54511)	2026-06-28 22:52:59 -07:00
insights.py
iteration_budget.py
jiter_preload.py
learn_prompt.py	fix(learn): honor requirements mixed with sources in /learn requests (#55956)	2026-06-30 16:56:01 -07:00
learning_graph.py	fix(learning_graph): guard non-dict metadata so /journey can't crash	2026-07-01 16:25:48 -05:00
learning_graph_render.py	fix(journey): swap skill/memory inks so drillable rows read as clickable	2026-06-30 11:54:16 -05:00
learning_mutations.py	refactor(journey): route memory mutations through MemoryStore atomic I/O	2026-06-30 15:16:21 -05:00
lmstudio_reasoning.py
manual_compression_feedback.py
markdown_tables.py
memory_manager.py	fix(agent): validate context/memory tool schemas before wrapping	2026-06-25 02:17:29 +05:30
memory_provider.py	fix(backup): capture memory-provider state stored outside HERMES_HOME (#50325)	2026-06-21 12:03:46 -07:00
message_content.py	fix(openviking): preserve structured sync attribution	2026-06-19 15:23:41 +08:00
message_sanitization.py	fix(agent): close tool-call sequence on all interrupt aborts, not just finalize_turn	2026-06-25 12:24:34 -05:00
moa_loop.py	feat(moa): add reference_max_tokens to cap advisor output and cut turn latency (#56756)	2026-07-02 00:16:35 -07:00
moa_trace.py	fix(moa): capture streamed aggregator output into full-turn traces (#56312)	2026-07-01 04:07:46 -07:00
model_metadata.py	fix(context): parse vLLM's token-based output-cap error format	2026-07-01 03:17:48 -07:00
models_dev.py
moonshot_schema.py
nous_rate_guard.py
onboarding.py
oneshot.py	feat(agent): one-shot LLM helper + llm.oneshot gateway RPC (#51261)	2026-06-23 08:01:50 +00:00
plugin_llm.py
portal_tags.py
process_bootstrap.py	fix(agent): honor custom CA certs on aux client + harden TLS resolution	2026-07-02 04:51:56 +05:30
prompt_builder.py	fix(agent): limit .hermes.md parent walk to git repos only	2026-06-28 20:46:32 -07:00
prompt_caching.py
rate_limit_tracker.py
reasoning_timeouts.py	fix(agent): detect thinking-timeout for reasoning models and surface actionable guidance instead of misleading file-write advice	2026-06-25 19:00:48 -07:00
redact.py	security(agent): redact Slack App-Level (xapp-) tokens	2026-07-01 02:45:22 -07:00
replay_cleanup.py	fix(tui): sanitize replay history on WebUI/TUI session resume (#29086) (#53939)	2026-06-27 20:56:49 -07:00
retry_utils.py	fix: handle named custom providers and Z.AI overload retries	2026-06-25 00:17:17 -07:00
runtime_cwd.py
secret_scope.py	feat(gateway): multiplex phase 2 — fail-closed profile credential isolation (Workstream A)	2026-06-19 07:34:15 -07:00
shell_hooks.py	feat(agent): add pre_verify hook and verify-on-stop coding guidance	2026-06-30 00:59:29 -05:00
skill_bundles.py
skill_commands.py
skill_preprocessing.py	fix(windows): hide console-window flash on backend git/gh/wmic/bash subprocess spawns	2026-06-28 05:28:45 -07:00
skill_utils.py	fix(curator): protect external skills from background curation	2026-06-25 22:03:02 -07:00
ssl_guard.py
ssl_verify.py	fix(agent): honor custom CA certs on aux client + harden TLS resolution	2026-07-02 04:51:56 +05:30
stream_diag.py
subdirectory_hints.py	fix(subdirectory_hints): catch RuntimeError from Path.expanduser()	2026-07-01 04:55:15 -07:00
system_prompt.py	feat(computer_use): cross-platform cua-driver (macOS/Windows/Linux)	2026-06-22 06:42:30 -07:00
think_scrubber.py
thinking_timeout_guidance.py	fix(agent): detect thinking-timeout for reasoning models and surface actionable guidance instead of misleading file-write advice	2026-06-25 19:00:48 -07:00
thread_scoped_output.py	fix(bg-review): scope stdout/stderr silencing to the worker thread (#55966)	2026-06-30 17:28:33 -07:00
title_generator.py	fix(title_generator): strip think blocks from LLM output before extracting title	2026-07-01 04:18:48 -07:00
tool_dispatch_helpers.py	fix(agent): wrap list-type untrusted content in untrusted_tool_result	2026-07-01 02:44:09 -07:00
tool_executor.py	fix(agent): prefer late-completing real result over timeout message (review)	2026-07-01 14:56:52 +05:30
tool_guardrails.py
tool_result_classification.py
trajectory.py
transcription_provider.py
transcription_registry.py
tts_provider.py
tts_registry.py
turn_context.py	fix(memory): degrade gracefully after repeated at-capacity consolidation failures (#42405)	2026-06-30 20:01:16 +05:30
turn_finalizer.py	fix(agent): persist recovered final responses	2026-07-01 03:34:49 -07:00
turn_retry_state.py	feat(vertex): add Google Vertex AI provider for Gemini (OAuth2)	2026-07-01 05:25:33 -07:00
usage_pricing.py	feat(vertex): add Google Vertex AI provider for Gemini (OAuth2)	2026-07-01 05:25:33 -07:00
verification_evidence.py	feat(agent): recognize focused ad-hoc verification scripts	2026-06-24 23:03:45 -05:00
verification_stop.py	feat(agent): restore surface-aware "auto" default for verify_on_stop	2026-06-30 01:43:08 -05:00
verify_hooks.py	feat(agent): add pre_verify hook and verify-on-stop coding guidance	2026-06-30 00:59:29 -05:00
vertex_adapter.py	security(vertex): route credential/project/region resolution through the profile secret scope	2026-07-02 06:07:56 +05:30
video_gen_provider.py
video_gen_registry.py
web_search_provider.py
web_search_registry.py