ismell0992-afk 3e99964789 fix(agent): prefer Ollama Modelfile num_ctx over GGUF training max ...

_query_local_context_length was checking model_info.context_length
(the GGUF training max) before num_ctx (the Modelfile runtime override),
inverse to query_ollama_num_ctx. The two helpers therefore disagreed on
the same model:

  hermes-brain:qwen3-14b-ctx32k     # Modelfile: num_ctx 32768
  underlying qwen3:14b GGUF         # qwen3.context_length: 40960

query_ollama_num_ctx correctly returned 32768 (the value Ollama will
actually allocate KV cache for). _query_local_context_length returned
40960, which let ContextCompressor grow conversations past 32768 before
triggering compression — at which point Ollama silently truncated the
prefix, corrupting context.

Swap the order so num_ctx is checked first, matching query_ollama_num_ctx.
Adds a parametrized test that seeds both values and asserts num_ctx wins.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

2026-04-13 04:24:07 -07:00

..

__init__.py

test: add unit tests for 8 modules (batch 2)

2026-02-26 13:54:20 +03:00

test_anthropic_adapter.py

fix: MiniMax/Alibaba incorrectly detected as Anthropic OAuth, causing mcp_ tool prefix (#7509)

2026-04-11 00:43:01 -07:00

test_auxiliary_client.py

fix: three provider-related bugs (#8161, #8181, #8147) (#8243)

2026-04-12 01:44:18 -07:00

test_auxiliary_config_bridge.py

refactor(tests): re-architect tests + fix CI failures (#5946)

2026-04-07 17:19:07 -07:00

test_auxiliary_named_custom_providers.py

fix(model): tighten direct-provider fallback normalization

2026-04-10 05:52:45 -07:00

test_compress_focus.py

feat: /compress <focus> — guided compression with focus topic (#8017)

2026-04-11 19:23:29 -07:00

test_context_compressor.py

fix(agent): route compression aux through live session runtime

2026-04-12 01:34:52 -07:00

test_context_engine.py

feat: wire context engine plugin slot into agent and plugin system

2026-04-10 19:15:50 -07:00

test_context_references.py

fix(agent): preserve quoted @file references with spaces

2026-04-10 13:05:01 -07:00

test_credential_pool_routing.py

refactor(tests): re-architect tests + fix CI failures (#5946)

2026-04-07 17:19:07 -07:00

test_credential_pool.py

fix(auth): gate Claude Code credential seeding behind explicit provider config

2026-04-10 05:19:21 -07:00

test_crossloop_client_cache.py

refactor(tests): re-architect tests + fix CI failures (#5946)

2026-04-07 17:19:07 -07:00

test_display_emoji.py

feat(tools): centralize tool emoji metadata in registry + skin integration

2026-03-15 20:21:21 -07:00

test_display.py

refactor(tests): re-architect tests + fix CI failures (#5946)

2026-04-07 17:19:07 -07:00

test_error_classifier.py

fix: add Alibaba/DashScope rate-limit pattern to error classifier

2026-04-10 05:52:45 -07:00

test_external_skills.py

feat(skills): support external skill directories via config (#3678)

2026-03-29 00:33:30 -07:00

test_insights.py

fix: remove 115 verified dead code symbols across 46 production files

2026-04-10 03:44:43 -07:00

test_local_stream_timeout.py

fix: is_local_endpoint misses Docker/Podman DNS names (#7950)

2026-04-11 14:46:18 -07:00

test_memory_provider.py

fix: remove 115 verified dead code symbols across 46 production files

2026-04-10 03:44:43 -07:00

test_memory_user_id.py

fix: thread gateway user_id to memory plugins for per-user scoping (#5895)

2026-04-07 11:14:12 -07:00

test_minimax_auxiliary_url.py

fix: provider/model resolution — salvage 4 PRs + MiniMax aux URL fix (#5983)

2026-04-07 22:23:28 -07:00

test_minimax_provider.py

fix: preserve dots in model names for OpenCode Zen and ZAI providers (#8794)

2026-04-12 21:22:59 -07:00

test_model_metadata_local_ctx.py

fix(agent): prefer Ollama Modelfile num_ctx over GGUF training max

2026-04-13 04:24:07 -07:00

test_model_metadata.py

fix: use ceiling division for token estimation, deduplicate inline formula

2026-04-11 16:33:40 -07:00

test_models_dev.py

fix: three provider-related bugs (#8161, #8181, #8147) (#8243)

2026-04-12 01:44:18 -07:00

test_prompt_builder.py

feat: add WSL environment hint to system prompt (#8285)

2026-04-12 02:26:28 -07:00

test_prompt_caching.py

fix(prompt-caching): skip top-level cache_control on role:tool for OpenRouter

2026-03-21 16:54:43 -07:00

test_rate_limit_tracker.py

feat: capture provider rate limit headers and show in /usage (#6541)

2026-04-09 03:43:14 -07:00

test_redact.py

test(redact): add regression tests for lowercase variable redaction (#4367) (#5185)

2026-04-05 00:10:16 -07:00

test_skill_commands.py

fix: sanitize Telegram command names to strip invalid characters

2026-04-06 11:27:28 -07:00

test_smart_model_routing.py

fix: hermes update causes dual gateways on macOS (launchd) (#1567)

2026-03-16 12:36:29 -07:00

test_subagent_progress.py

feat(api): structured run events via /v1/runs SSE endpoint

2026-04-05 12:05:13 -07:00

test_subdirectory_hints.py

fix(agent): catch PermissionError in subdirectory hint discovery

2026-04-09 03:10:30 -07:00

test_title_generator.py

feat: auto-generate session titles after first exchange

2026-03-17 04:14:40 -07:00

test_usage_pricing.py

feat: use endpoint metadata for custom model context and pricing (#1906)

2026-03-18 03:04:07 -07:00