Code
Hub
Workspaces
Following
Trending
Connect
MCP
copy
Create free account
hub
/
github.com/HKUDS/Vibe-Trading
/ functions
Functions
26,887 in github.com/HKUDS/Vibe-Trading
⨍
Functions
26,887
◇
Types & classes
2,943
↳
Endpoints
204
Function
test_a_shortlist_answers_the_user_but_still_cannot_fetch_a_quote
#955 closed half of this: the skill loaded, the answer stayed blocked.
agent/tests/test_agent_grounding.py:2413
Method
test_a_shortlist_candidate_is_allowed
A resolver that offered the symbol is a tool that returned it.
agent/tests/test_agent_output_discipline.py:635
Function
test_a_sign_before_the_currency_mark_is_the_figures_sign
( text: str, values: list[float] )
agent/tests/test_grounding_backtest_outputs.py:1057
Method
test_a_signal_inside_the_warmup_never_fills
(self, tmp_path)
agent/tests/test_warmup_window.py:236
Function
test_a_single_alpha_has_no_search_to_correct_for
(monkeypatch)
agent/tests/test_bench_multiple_testing.py:95
Function
test_a_single_deal_american_waterfall_matches_european_waterfall
A one-deal American waterfall must reduce exactly to a European one.
agent/tests/quantlib/test_fundmath.py:1012
Method
test_a_single_market_is_allowed
(self)
agent/tests/test_composite_currency_guard.py:61
Method
test_a_single_oversized_record_is_still_emitted
(self)
agent/tests/test_tool_result_paging.py:98
Function
test_a_single_period_group_keeps_its_decimal_reading
()
agent/tests/test_grounding_unit_and_scanner_gaps.py:158
Function
test_a_single_valuation_cannot_produce_a_return
An opening and a closing mark are both required.
agent/tests/quantlib/test_performance.py:300
Function
test_a_skipped_retrain_keeps_serving_the_previous_model
The retrain day itself is predicted too, not left at 0.0 (#1522).
agent/tests/test_ml_strategy_skill.py:131
Function
test_a_small_window_compacts_before_it_overflows
(tmp_path, monkeypatch)
agent/tests/test_context_budget.py:284
Method
test_a_smile_matches_the_smile_model
(self)
agent/tests/test_options_smile_consistency.py:103
Function
test_a_smoother_skips_the_gap
Skip and continue, not NaN until rewarmed: the policy set on #1463.
agent/tests/factors/test_missing_bar_fabrication.py:210
Function
test_a_source_only_the_note_names_is_no_citation
The block is stripped before release, so the reader must see the source.
agent/tests/test_grounding_role_hardening.py:184
Function
test_a_source_that_never_succeeded_reports_no_last_success_time
last_success_at is None rather than invented when no healthy snapshot exists.
agent/tests/test_portfolio_service.py:469
Function
test_a_source_without_a_selected_account_errors_before_any_call
(tmp_path, broker: _Broker)
agent/tests/test_robinhood_portfolio_source.py:104
Function
test_a_stand_in_reaches_no_model_written_file_and_nothing_outside_the_run
The model's own config.json stays non-evidence, and a sibling directory holding a metrics file of the same name is not the active run.
agent/tests/test_grounding_backtest_outputs.py:1009
Function
test_a_stray_opener_cannot_swallow_the_declaration_block
A stray "```python" used to pair with "```figures" and leave the block in the answer.
agent/tests/test_grounding_shape_hardening.py:273
Function
test_a_stray_quote_that_swallows_the_tail_is_refused
An unbalanced quote in the last column absorbs every following line.
agent/tests/test_extraetf_reader.py:563
Function
test_a_streamed_answer_never_shows_its_figures_block
A run with no evidence streams live, and the block still stays out.
agent/tests/test_grounding_release.py:2134
Function
test_a_strict_comparison_is_refused_instead_of_dropped
A strict operator has no exact inclusive render, so it errors instead.
agent/tests/test_shadow_codegen_condition_bounds.py:188
Function
test_a_string_false_invert_sign_leaves_file_amounts_alone
A JSON-string "false" must not be read as truthy and negate every row.
agent/tests/test_cashflow_analytics_tool.py:108
Method
test_a_sub_threshold_cover_page_is_rescaled_from_the_information_table
A filer still reporting thousands must not be shrunk 1000x and mis-ranked. Modelled on CapitalatWork S.A. (CIK 0002135126), whose 2026-03-31
agent/tests/test_institutional_holdings_tool.py:1402
Function
test_a_sub_year_window_is_not_annualised
Stating a full-year figure that was never observed is extrapolation.
agent/tests/quantlib/test_performance.py:459
Function
test_a_sum_of_squared_weights_is_arithmetic
(two_runs: GroundingLedger)
agent/tests/test_grounding_backtest_outputs.py:757
Function
test_a_summary_file_the_run_card_does_not_vouch_for_is_ignored
(tmp_path: Path)
agent/tests/test_grounding_backtest_outputs.py:372
Method
test_a_superseded_table_is_never_downloaded
Reading a table only to throw it away costs a full document fetch.
agent/tests/test_institutional_holdings_tool.py:815
Function
test_a_sweep_running_inside_another_sweep_cannot_double_send
The race the claim exists for: a second sweep while the first sends. A re-read alone would not have caught this — while the first send is in
agent/tests/test_scheduled_delivery_outbox.py:223
Method
test_a_symbol_a_succeeding_tool_accepted_is_allowed
Backtest configs name their universe before any bar is fetched.
agent/tests/test_agent_output_discipline.py:342
Function
test_a_symbol_the_declaration_names_outranks_the_prose
(tmp_path: Path)
agent/tests/test_grounding_role_hardening.py:474
Method
test_a_symbol_the_user_supplied_is_allowed
(self, tmp_path: Path)
agent/tests/test_agent_output_discipline.py:361
Method
test_a_symbol_with_no_factors_is_dropped_not_returned_raw
(self)
agent/tests/test_tushare_loader.py:234
Function
test_a_table_cell_and_the_same_prose_figure_get_one_verdict
Table and prose are the same surface now, so they cannot disagree. The generic-header fallback existed to keep a table from being looser than
agent/tests/test_agent_grounding.py:3484
Function
test_a_table_grounds_only_the_rows_the_model_was_shown
(two_runs: GroundingLedger)
agent/tests/test_grounding_backtest_outputs.py:303
Method
test_a_table_listed_after_the_decoys_is_still_probed_first
(self)
agent/tests/test_institutional_holdings_tool.py:1566
Method
test_a_table_row_counts_as_attaching_a_figure
(self, tmp_path: Path)
agent/tests/test_agent_output_discipline.py:266
Function
test_a_tail_risk_figure_a_tool_returned_is_grounded
(tmp_path: Path)
agent/tests/test_grounding_role_hardening.py:586
Function
test_a_tail_risk_ref_grounds_its_own_value
(tmp_path: Path, ref: str)
agent/tests/test_grounding_role_hardening.py:631
Function
test_a_tampered_stored_flag_is_recomputed_not_trusted
()
agent/tests/test_principal.py:59
Function
test_a_text_answer_that_crosses_the_goal_budget_is_released
(tmp_path, monkeypatch)
agent/tests/test_llm_cost_accounting.py:357
Function
test_a_thousands_group_longer_than_three_digits_is_refused
``1234.567`` is a US decimal; reading its period as German scales it 1000x. The leading group is the one the grouping rule used to skip, and it i
agent/tests/test_extraetf_reader.py:808
Function
test_a_thousands_grouping_stays_grouped
(text: str, values: list[float])
agent/tests/test_grounding_shape_hardening.py:234
Function
test_a_three_character_currency_word_is_money_and_a_four_character_run_is_not
( tmp_path: Path, )
agent/tests/test_grounding_shape_hardening.py:660
Function
test_a_ticker_ending_in_a_is_not_a_market_constraint
(tmp_path: Path, message: str)
agent/tests/test_grounding_resolution_context.py:318
Function
test_a_tier_that_does_not_conserve_cash_is_rejected_at_construction
()
agent/tests/quantlib/test_fundmath.py:579
Function
test_a_tighter_signal_threshold_flags_more_rows
()
agent/tests/quantlib/test_credit.py:415
Function
test_a_timestamp_label_cannot_end_before_it_starts
The positional branch rejected this; the Series branch clipped it to its start.
agent/tests/quantlib/test_crossvalidation.py:473
Function
test_a_tool_scoped_ref_with_a_nonexistent_value_stays_rejected
(tmp_path: Path)
agent/tests/test_grounding_container_field_refs.py:213
Method
test_a_top_n_capped_list_does_not_pass_itself_off_as_complete
``paging.complete`` covers the ranked slice, not the whole portfolio.
agent/tests/test_institutional_holdings_tool.py:515
Method
test_a_top_n_cut_full_portfolio_says_it_was_cut
(self)
agent/tests/test_etf_holdings_tool.py:1074
Function
test_a_trade_price_read_back_is_not_a_market_print
An execution price is left out of the table evidence, or reading trades.csv would turn a backtest report into a price answer owing a source and cu
agent/tests/test_grounding_backtest_outputs.py:701
Function
test_a_trade_within_capacity_fills_completely
()
agent/tests/test_factor_costs.py:28
Method
test_a_trading_halt_does_not_change_the_verdict
The median reports the regular spacing; one long gap cannot outvote it.
agent/tests/test_cross_market_annualization.py:218
Function
test_a_transactions_export_refuses_to_pose_as_a_portfolio
()
agent/tests/test_extraetf_reader.py:259
Function
test_a_transient_failure_survives_one_identical_retry
The case the block is NOT for: a read that fails once for a reason that has nothing to do with its arguments. Only a successful *mutating* ca
agent/tests/test_agent_loop_no_progress.py:203
Function
test_a_trial_plan_inside_the_capital_fit_is_not_a_finding
A sleeve the search scales down to a fill leaves no rejection at all.
agent/tests/test_plan_rejected.py:382
Function
test_a_true_decimal_comma_close_passes_against_its_print
(tmp_path: Path)
agent/tests/test_grounding_shape_hardening.py:238
Function
test_a_two_sign_change_schedule_reports_both_roots
The textbook multiple-IRR pattern: out, in, out again. Two rates make the net present value zero, so reporting one of them as "the" IRR witho
agent/tests/quantlib/test_fundmath.py:284
Function
test_a_typo_still_raises_rather_than_being_guessed_at
(raw)
agent/tests/quantlib/test_options.py:475
Function
test_a_unicode_minus_is_a_sign_and_an_en_dash_is_a_range
()
agent/tests/test_grounding_shape_hardening.py:147
Function
test_a_unit_character_that_starts_a_word_is_not_swallowed
0.95 元宵节" must not become "(略※)宵节".
agent/tests/test_grounding_release.py:1519
Function
test_a_us_csv_stem_resolves_to_its_venue_suffix
``INTC_US.csv`` is ``INTC.US``; without the row it was no evidence at all.
agent/tests/test_agent_grounding.py:2667
Method
test_a_us_suffixed_query_still_matches_the_ticker
(self)
agent/tests/test_etf_holdings_tool.py:530
Method
test_a_us_suffixed_symbol_still_reaches_the_sec_ticker
(self)
agent/tests/test_etf_holdings_tool.py:453
Function
test_a_v1_1_call_keeps_the_frozen_document
()
agent/tests/test_paper_trade_5m_tool.py:122
Function
test_a_v1_1_document_still_has_no_engine_fingerprint
()
agent/tests/test_paper_trade_1d_fingerprint.py:86
Function
test_a_v1_2_field_silently_downgraded_to_v1_1_shows_up_only_in_the_hash
A v1.2 payload sent to a v1.1 parser runs the daily path, but the hash differs. The v1.1 parser has no fields for the four selectors, so it ignor
agent/tests/test_paper_trade_5m_tool.py:136
Function
test_a_value_error_stays_a_non_retryable_invalid_request
(monkeypatch)
agent/tests/test_paper_trade_5m_tool.py:194
Function
test_a_value_is_called_repeated_only_from_its_second_rejection
The current draft is not its own predecessor. Comparing against every recorded validation, the current one included, told the model on its FI
agent/tests/test_grounding_figures_contract.py:639
Function
test_a_value_that_does_not_exist_gets_no_candidate_and_stays_rejected
( tmp_path: Path, )
agent/tests/test_grounding_container_field_refs.py:106
Function
test_a_value_that_rounds_the_other_way_is_rejected
30.205165 rounds to 30,21; 30,20 is 0.0052 away, outside half a unit of two decimals.
agent/tests/test_grounding_numeric_locale_rounding.py:294
Function
test_a_value_that_would_change_on_the_way_to_a_float_is_refused
``float`` holds about seventeen significant digits, so a value past that would arrive with a different magnitude even though it quantized cleanly.
agent/tests/test_extraetf_reader.py:837
Function
test_a_value_the_file_does_not_hold_is_refused_under_every_stand_in
( own_run: GroundingLedger, spelling: str )
agent/tests/test_grounding_backtest_outputs.py:976
Method
test_a_vietnam_rule_engine_is_built
(self)
agent/tests/test_composite_vietnam_rules.py:68
Method
test_a_vn_book_cannot_run_beside_another_currency
(self)
agent/tests/test_composite_vietnam_rules.py:155
Method
test_a_vwap_without_volume_is_unknown_not_the_last_day
(self)
agent/tests/test_weekly_monthly_bars.py:72
Method
test_a_week_and_a_month_are_calendar_counts
(self, source)
agent/tests/test_weekly_monthly_bars.py:177
Method
test_a_week_runs_monday_to_sunday_and_is_stamped_with_its_last_trading_day
(self)
agent/tests/test_weekly_monthly_bars.py:37
Method
test_a_weekly_file_is_annualised_as_weekly
A weekly file declared finer is read as ``1W``: 52 a year. Keeping the declaration was the bug -- a weekly file declared ``1H`` annua
agent/tests/test_cross_market_annualization.py:268
Function
test_a_weight_table_changes_the_composite_over_time
()
agent/tests/quantlib/test_combination.py:141
Function
test_a_weight_table_may_cover_only_rebalance_dates
()
agent/tests/quantlib/test_combination.py:157
Function
test_a_window_smaller_than_the_static_prompt_still_has_a_floor
()
agent/tests/test_context_budget.py:121
Function
test_a_window_too_small_for_the_summary_tail_does_not_resummarise_every_turn
80K window: static ~48K + the ~20K tail a summary keeps is over the 64K line, so without the regrowth rule every iteration paid for a summary.
agent/tests/test_context_budget.py:368
Function
test_a_worded_cell_still_checks_an_integer_price_in_the_thousands
#1471 reads "1520 (limit)" as prose, and prose checks an integer price of 600519.SH.
agent/tests/test_grounding_role_hardening.py:1150
Function
test_a_write_makes_earlier_readonly_results_unreplayable
A result cached before a write may describe a file the write changed: after the write, a lost call runs again and reads the new file.
agent/tests/test_readonly_replay_after_microcompact.py:518
Function
test_a_wrong_argument_returns_the_signature_to_fix_it_with
(tool)
agent/tests/test_quantlib_tool.py:212
Function
test_a_wrong_call_ref_hints_only_refs_that_then_ground_the_figure
The hint for a backtest's value names the call that wrote it, and each ref it lists validates.
agent/tests/test_grounding_backtest_outputs.py:270
Function
test_a_wrong_long_precision_spanish_percentage_is_still_rejected
( tmp_path: Path, )
agent/tests/test_grounding_numeric_locale_rounding.py:211
Function
test_a_yahoo_index_is_raw
Yahoo's adjclose equals close on ^GSPC / ^NDX / ^HSI: nothing was adjusted.
agent/tests/test_price_caliber.py:204
Function
test_a_year_shaped_number_with_a_measurement_mark_is_measured
(text: str, figure: str)
agent/tests/test_grounding_shape_hardening.py:311
Function
test_a_year_under_a_price_column_is_checked
(tmp_path: Path)
agent/tests/test_grounding_shape_hardening.py:346
Function
test_a_zero_hurdle_accrues_nothing
()
agent/tests/quantlib/test_fundmath.py:651
Function
test_aapl_ratio_change_books_no_artificial_loss
(tmp_path, monkeypatch)
agent/tests/test_argentina_backtests.py:216
Function
test_aapl_without_the_ratio_change_fails_closed
(tmp_path, monkeypatch)
agent/tests/test_argentina_backtests.py:249
Function
test_abbreviate_num_unit_boundary
(value: float, expected: str)
agent/tests/test_cli_format_unit_boundaries.py:24
Function
test_abbreviated_fx_pair_x_classes_as_forex
JPY=X (USD/JPY abbreviation) must not fall to the a_share default.
agent/tests/test_market_data.py:113
Method
test_above_no_arbitrage_ceiling_raises
(self)
agent/tests/quantlib/test_options.py:257
Method
test_absent_declaration_evaluates_everything
(self)
agent/tests/test_warmup_window.py:69
Method
test_absent_effort_is_dropped_from_the_request
None must not serialize as a null field on unsupported providers.
agent/tests/test_llm_reasoning_effort.py:628
← previous
next →
14,401–14,500 of 26,887, ranked by callers