Knowledge Graph Report — llm-mailroom

DocumentState, DocumentManifest, PipelineStage, apply_intake(), arbiter_node(), archive_node(), boss_escalation_node(), _build_checkpointer() (+82 more)

_denial_reasons(), determination_consistency_is_quality(), honesty_trace_metadata(), insurance_determination_consistent(), insurance_determination_issues(), insurance_expected_set_is_homogeneous(), insurance_gt_is_homogeneous(), _norm_determination() (+75 more)

FakeLangChainLLM, _FakeStructuredRunner, chat, _Choices, completions, _EvalLangChainLLM, _HintedEvalLangChainLLM, _MockClient (+51 more)

get_queue(), _move_rejected_to_failed(), _ensure_dirs(), accepted_extensions(), archive_dir(), classified_dir(), clear_ingestion_paused(), failed_dir() (+50 more)

ArbiterAgent, BossAgent, ComplianceSpecialist, CorporateRecordsSpecialist, CorrespondenceSpecialist, InsuranceClaimsSpecialist, .arbitrate(), .system_prompt() (+48 more)

configure(), flush_braintrust(), instrument_openai_client(), is_configured(), _apply_taxonomy_settings(), client_kwargs(), flush_langfuse(), get_trace_id() (+47 more)

after_arbiter(), after_boss(), after_classify(), after_extraction(), after_extraction_gated(), after_human_review(), after_judge(), after_report() (+46 more)

AuditLogRecord, DocumentRecord, MatterRecord, Base, _all_chains(), main(), get_audit_chain(), get_latest_audit_hash() (+41 more)

SorterReviewerAgent, SorterAgent, .review(), .system_prompt(), build_structured_schema(), format_sorter_subclass_catalogs(), sorter_subclass_catalog(), valid_sorter_subclasses() (+33 more)

ComplianceFilingSpecialist, CorporateRecordsSpecialist, CorrespondenceSpecialist, _SpecialistBase, get_prompt(), .system_prompt(), .system_prompt(), .system_prompt() (+27 more)

BaseAgent, .augmented_system_prompt(), ._build_user_content(), ._call_llm(), ._call_structured(), ._call_vision(), ._call_vision_multi(), ._check_deadline() (+25 more)

_check_database(), get_audit_trail(), get_document_status(), get_matter(), health(), lifespan(), ops_resume(), ops_status() (+25 more)

clause_handoff(), skip_conflict_field(), coerce_gt_value(), _compact(), enrich_extraction(), _normalize(), normalize_claim_type(), normalize_communication_type() (+25 more)

RunBudgetExceeded, RunDeadlineExceeded, _bounded(), compute_run_metrics(), check_token_budget(), estimate_cost(), get_call_timeout_seconds(), get_deadline_seconds() (+20 more)

classify_image(), clean_prediction(), extract_confidence(), extract_reasoning(), extract_runner_up(), _valid_classes(), build_text_messages(), build_vision_messages() (+17 more)

_resolved_models(), agent_uses_vision(), _any_specialist_uses_vision(), is_vision_capable(), max_pages(), pipeline_uses_vision(), render_document_pages(), render_image() (+16 more)

ensure_field_score_configs(), score_and_log_extraction(), _client(), emit_pipeline_scores(), ensure_score_configs(), is_enabled(), langfuse_score_name(), _score_data_type() (+16 more)

ImageExtractor, extract_text_from_image(), .extract(), ._extract_with_vision(), ._fallback_extract(), .system_prompt(), ._llm_transcribe(), compile_matter_record() (+15 more)

RunResult, build_parser(), main(), log_run(), _model_name(), print_summary(), run_task(), _tokens_summary() (+14 more)

_build_evaluator_request(), _build_output_definition(), _build_rule_request(), _client(), _current_evaluator_prompt(), _ensure_llm_connection(), _existing_rule_ids(), main() (+14 more)

append_record(), build_record(), default_log_path(), default_sibling_root(), git_snapshot(), _inside(), regenerate(), _run_python() (+13 more)

get_field_types(), warm_embedding_model(), main(), _perturb_date(), _perturb_entity_list(), _perturb_free_text(), _perturb_money(), _perturb_name() (+13 more)

_contracts_from_annotations(), _contracts_from_txt(), _download(), download_all(), _list_hf_files(), _load_subtype_taxonomy(), main(), _normalize_category() (+13 more)

_attach_field_scoring(), diff_report(), filter_real_samples(), _ground_truth_scores(), _ingest_scores(), main(), misfile_candidates(), _parse_expected_fields() (+12 more)

InboxHandler, Watcher, claim_file(), ._infer_matter_id(), .__init__(), ._is_processable(), .on_created(), ._process() (+11 more)

ProviderConfig, .__init__(), _check_llm_provider(), get_llm(), get_llm_client(), get_llm_model(), instrument_client(), _build_providers() (+10 more)

as_clause_lines(), enrich_contract_extraction(), flatten_cuad_clause_labels(), flatten_maud_clause_labels(), infer_merger_consideration(), normalize_consideration(), parse_json_obj(), _as_meta() (+10 more)

ContractsSpecialist, ContractsSpecialist, .__init__(), .__init__(), .system_prompt(), get_extraction_schema(), get_all_doc_types(), get_doc_class() (+9 more)

CorpusUnavailable, Sample, _fingerprint(), load_cuad_qa(), load_family_rows(), _normalize_prediction(), _extract_binary(), data.py (+9 more)

_is_retryable_error(), _is_json_mode_400(), _is_retryable(), _retry_after_seconds(), _retry_config(), retry_sleep_seconds(), _status_code(), check_run_deadline() (+9 more)

OpsMonitor, _main(), ._analyze_metrics(), ._gather_metrics(), .__init__(), .is_paused(), .pause_info(), ._query_catalog() (+8 more)

AuditLogEntry, archive_document(), _file_sha256(), build_audit_entry(), compute_audit_hash(), compute_audit_hash_v1(), verify_chain(), _verify() (+8 more)

LegalBenchAgent, .answer_binary(), .augmented_system_prompt(), .classify_family(), .__init__(), .system_prompt(), .usage(), agent.py (+8 more)

attach_run_scores(), ensure_score_configs_if_enabled(), _environment(), legalbench_trace(), question_observation(), is_enabled(), pipeline_trace(), langfuse_tracing.py (+8 more)

AgentTool, ._tool_context(), .__init__(), .run(), _build_toolkit(), get_tools(), render_tools(), _tool_field_types() (+7 more)

_NoopLangfuse, _NoopSpan, .create_trace_id(), .flush(), .get_current_trace_id(), .set_current_trace_io(), .shutdown(), .start_as_current_observation() (+7 more)

attach_single_doc_extras(), _numeric_extra(), score_and_log_intake(), score_intake_suite(), score_with_suite(), unwrap_suite_result(), suite_scoring.py, Any (+7 more)

default_environment(), load_env(), _client(), _cost_models(), _existing_by_name(), main(), _match_pattern(), _prices_match() (+7 more)

apply_classification_guard(), apply_extraction_guard(), guard_classification(), guard_extraction(), _has_substantive_content(), _is_valid_confidence(), _valid_subtypes(), guards.py (+7 more)

_escape(), generate_pdf_from_text(), _load_manifest(), prepare_samples(), _client(), _doc_text(), _ensure_dataset(), main() (+7 more)

BaseAgent, ._build_multimodal(), ._call_llm(), ._call_structured(), ._configured_max_input_chars(), ._configured_max_tokens(), ._configured_reasoning_effort(), .system_prompt() (+6 more)

ComplianceFilingExtraction, ContractExtraction, CorporateRecordExtraction, CorrespondenceExtraction, InsuranceClaimExtraction, Matter, get_extraction_schema(), judge.py (+6 more)

WidgetSpec, _client(), _existing_placements(), json_dumps(), main(), _placement_kwargs(), _score_widget(), _spec_to_request() (+6 more)

main(), _slug(), _client(), main(), _parse_since(), sync_logs(), _trace_basics(), _trace_stage() (+6 more)

equivalent_subtypes(), _binary_f1(), _ece(), _mean(), _safe_div(), score_binary(), score_multiclass(), scoring.py (+6 more)

_RotatingFileSink, .__call__(), .__init__(), setup_logging(), _base_env(), main(), run_config(), RotatingFileHandler (+4 more)

CompletenessJudge, ._field_list(), .judge_classification(), .judge_completeness(), .judge_extraction_correctness(), .system_prompt(), ._taxonomy_spec(), ._truncate() (+4 more)

_memory_dir(), _memory_path(), recent_context(), record_outcome(), stats(), _tool_memory(), memory.py, Path (+4 more)

list_prompts(), PROMPT_TEMPLATES(), docclass_prompts_enabled(), langchain_prompt_version(), managed_prompt_lookup(), langchain_agents/prompts.py, docclass_mode.py, List all available prompt versions. (+4 more)

_date_pair_days(), extraction_diagnostics(), _mean(), _median(), parse_duration_days(), _r2(), metrics.py, Run-level diagnostic metrics for extraction scoring. Ported from ``llm-entity-… (+4 more)

MockLegalBenchModel, _hash(), .answer_binary(), .classify_family(), .__init__(), .last_usage(), .usage(), legalbench/mock.py (+3 more)

_latest_audit_hash(), _persist_provenance(), _persist_scores(), _run_coro(), _touch_heartbeat(), _write_catalog_record(), Run a coroutine from a sync context: schedule it on the running loop when one…, Best-effort fetch of the last entry_hash for this doc_id (the previous link of… (+3 more)

bootstrap_ci(), _clean(), delta_significance(), _resample_means(), bootstrap.py, Any, Random, Bootstrap confidence intervals and small-sample delta testing. Ported verbatim… (+3 more)

create_trace_score(), is_real_sample(), _dim_summary(), _ingest(), judge_one(), main(), print_summary(), _raw_text_for() (+3 more)

_caption_from_text(), _download(), fetch_atticus(), fetch_legalbench(), fetch_pileoflaw(), main(), _stream_pol_records(), fetch_external_samples.py (+3 more)

_langchain_prompt(), prompt_templates(), get_langfuse_client(), _client(), _current_production(), main(), sync_one(), sync_prompts.py (+2 more)

cutover_agent(), cutover_all(), list_agents(), list_local_models(), load_config(), main(), recommend_cutover_order(), save_config() (+2 more)

PDFTranscriber, ._extract_raw_text(), ._looks_clean_text_pdf(), .system_prompt(), .transcribe(), transcribe_pdf(), BaseAgent, Path (+1 more)

LegalBenchTask, _binary_classes(), _call_binary(), _call_family(), _extract_family(), _family_classes(), _family_labels(), tasks.py (+1 more)

bool_env(), get_env(), load_env(), require_env(), env_utils.py, Load ``braintrust.env`` then ``.env`` into the environment (idempotent).…, Validate that all given environment variables are set and non-empty. Returns…, Get an environment variable with a default fallback. (+1 more)

_append(), _build_versions(), _rules(), _specialist_rules(), prompts_docclass.py, Docclass prompt variants for every mailroom classification-chain role.…, Pure-appended docclass variant: base is a STRICT PREFIX of the result., Derive every variant from the live production template of that role. (+1 more)

build_report(), _clean_extracted(), _field_score_for(), _fmt_usd(), _json_block(), _load_config(), main(), _manifest_rows() (+1 more)

SorterAgent, .classify(), .classify_json(), .__init__(), _LangChainSorterAgent, Mailroom-configured sorter. - Model/budget defaults come from ``taxonomy.yaml``…, Classify a document, optionally with page images attached. Returns ``(doc_type,…, Structured classify used by the graph (includes ``doc_subclass``).

_check_cost_watchdog(), _fetch_openrouter_prices(), _price_for(), _record_langchain_response(), Warn at $0.15, abort the run at $0.20 (cumulative across all samples)., Mirror _wrap_client's usage/cost accounting for a LangChain response., Fetch live OpenRouter pricing (per-token), normalized to $/M tokens. The…

family_classification_prompt_v1(), get_prompt(), legalbench/prompts.py, Versioned LegalBench task prompts. Prompt version = experiment identity in the…, Fill the 25-family list into the multiclass prompt (called per run so the…, Resolve a prompt version to its system-prompt text.

_aggregate(), _cell(), main(), _print_table(), _scores_of(), compare_runs.py

_chunk_config(), _extract_contracts(), _run_chunked_extraction(), Chunked-extraction config from taxonomy.yaml (`chunking:` block). Chunking…, Run a specialist extraction, chunking long documents (v15+ pass).…

_prompt_versions(), _bound_prompt_versions(), Prompt versions bound during the run (best-effort; Langfuse-managed prompts…, Version keys currently wired into production / agent defaults. Used for catalog…

field_is_ambiguous(), get_type_bands(), Per-field-type ambiguous-band overrides from ``field_scoring.type_bands``.…, Is this field score in the (possibly type-specific) ambiguous band? Band check…

_require_token(), Request, Dependency: reject requests without the bearer token (audit L-2).