Teknium e5af1dd633 fix(review): tell background reviewer not to capture transient env failures as skills (#23004) ... Closes #6051. Reported failure mode: agent migrated to WSL2, browser launch failed because Playwright wasn't installed yet. Background reviewer captured the failure as a durable skill (`browser-tool-launch-issue`) and the agent kept refusing the browser tool for weeks after Playwright was installed and verified working. Negative claims also propagated into unrelated skills ("browser tools do not work", "cannot use Y from execute_code"). Root cause: `_SKILL_REVIEW_PROMPT` and `_COMBINED_REVIEW_PROMPT` both lean hard on "be active, save things, a pass that does nothing is a missed learning opportunity." Neither distinguished durable knowledge from transient environment state. The reviewer was doing what it was told. Fix at the write site — both prompts now carry a "Do NOT capture" section calling out: • Environment-dependent failures (missing binaries, fresh-install errors, post-migration path mismatches, 'command not found', unconfigured credentials, uninstalled packages) • Negative claims about tools or features ("X does not work") that harden into self-cited refusals • Session-specific transient errors that resolved before the conversation ended • One-off task narratives ("summarize today's market", "analyze this PR") — also addresses the #12812 / #4538 family Plus a positive-reframing line: when a tool fails because of setup state, capture the FIX (install command, config step, env var) under an existing setup/troubleshooting skill — never "this tool doesn't work" as a standalone constraint. Targeted tests: 24/24 passing in tests/run_agent/test_review_prompt_class_first.py (2 new + all existing review-prompt assertions). Substring-based checks so future prompt edits don't false-fail.		2026-05-09 22:51:25 -07:00
..
__init__.py
conftest.py	test: speed up slow tests (backoff + subprocess + IMDS network) (#11797)	2026-04-17 14:21:22 -07:00
test_413_compression.py	fix(agent): surface preflight compression status	2026-05-04 01:41:51 -07:00
test_860_dedup.py	fix: lazy session creation — defer DB row until first message (#18370)	2026-05-01 18:39:12 +05:30
test_1630_context_overflow_loop.py	fix(tests): make AIAgent constructor calls self-contained (#11755)	2026-04-17 12:32:03 -07:00
test_agent_guardrails.py	fix(agent): include name field on every role:tool message for Gemini compatibility (#16478)	2026-05-04 05:06:33 -07:00
test_agent_loop.py
test_agent_loop_tool_calling.py
test_agent_loop_vllm.py
test_anthropic_error_handling.py	feat(providers): extend request_timeout_seconds to all client paths	2026-04-19 11:23:00 -07:00
test_anthropic_prompt_cache_policy.py	fix(minimax): enable Anthropic prompt caching for MiniMax's own models (#17425)	2026-04-29 04:56:55 -07:00
test_anthropic_third_party_oauth_guard.py	fix(anthropic): complete third-party Anthropic-compatible provider support (#12846)	2026-04-19 22:43:09 -07:00
test_anthropic_truncation_continuation.py	refactor: remove _nr_to_assistant_message shim + fix flush_memories guard	2026-04-23 02:30:05 -07:00
test_api_max_retries_config.py	feat(agent): make API retry count configurable via agent.api_max_retries (#14730)	2026-04-23 13:59:32 -07:00
test_async_httpx_del_neuter.py	fix(copilot): send vision header for Copilot vision requests	2026-04-27 08:35:50 -07:00
test_background_review.py	fix(cli): surface self-improvement review summaries from bg thread	2026-04-30 14:07:22 -07:00
test_background_review_summary.py	fix(agent): exclude prior-history tool messages from background review summary	2026-04-24 03:10:19 -07:00
test_background_review_toolset_restriction.py	fix(ci): stabilize main test suite regressions (#17660)	2026-04-29 23:18:55 -07:00
test_codex_multimodal_tool_result.py	feat(vision): vision_analyze returns pixels to vision-capable models, not aux text (#22955)	2026-05-09 21:06:19 -07:00
test_commit_memory_session_context_engine.py	fix(agent): notify context engine on commit_memory_session (#22764)	2026-05-09 12:28:42 -07:00
test_compress_focus_plugin_fallback.py	refactor(memory): remove flush_memories entirely (#15696)	2026-04-25 08:21:14 -07:00
test_compression_boundary.py
test_compression_boundary_hook.py	fix: signal compression boundary to context engine	2026-04-26 19:07:18 -07:00
test_compression_feasibility.py	refactor(memory): remove flush_memories entirely (#15696)	2026-04-25 08:21:14 -07:00
test_compression_persistence.py	fix(tests): make AIAgent constructor calls self-contained (#11755)	2026-04-17 12:32:03 -07:00
test_compression_trigger_excludes_reasoning.py	fix(compression): exclude completion tokens from compression trigger (#12026)	2026-04-20 05:12:10 -07:00
test_compressor_fallback_update.py
test_concurrent_interrupt.py	test: remove 50 stale/broken tests to unblock CI (#22098)	2026-05-08 14:55:40 -07:00
test_context_token_tracking.py	feat(providers): extend request_timeout_seconds to all client paths	2026-04-19 11:23:00 -07:00
test_copilot_native_vision_headers.py	fix(copilot): mark native image requests as vision	2026-04-27 08:35:50 -07:00
test_create_openai_client_kwargs_isolation.py	fix(tests): make AIAgent constructor calls self-contained (#11755)	2026-04-17 12:32:03 -07:00
test_create_openai_client_proxy_env.py	test(proxy): regression tests for NO_PROXY bypass on keepalive client	2026-04-24 03:04:42 -07:00
test_create_openai_client_reuse.py	fix(tests): make AIAgent constructor calls self-contained (#11755)	2026-04-17 12:32:03 -07:00
test_deepseek_reasoning_content_echo.py	fix(deepseek): use non-empty reasoning_content placeholder for V4 Pro thinking mode	2026-04-30 23:04:23 -07:00
test_deepseek_v4_thinking_live.py	fix(deepseek): preserve v4 reasoning_content on replay	2026-04-30 11:18:39 -07:00
test_dict_tool_call_args.py	fix(tests): fix 78 CI test failures and remove dead test (#9036)	2026-04-13 10:50:24 -07:00
test_empty_response_recovery_persistence.py	fix(run_agent): break permanent empty-response loop from orphan tool-tail (#21385)	2026-05-07 08:35:10 -07:00
test_exit_cleanup_interrupt.py	test: speed up slow tests (backoff + subprocess + IMDS network) (#11797)	2026-04-17 14:21:22 -07:00
test_fallback_model.py	fix(fallback): resolve api_key_env in fallback chain entries (carve-out of #22665)	2026-05-09 17:53:56 -07:00
test_image_rejection_fallback.py	fix(computer-use): harden image-rejection fallback + AUTHOR_MAP	2026-05-08 11:07:38 -07:00
test_image_shrink_recovery.py	feat(image-input): native multimodal routing based on model vision capability (#16506)	2026-04-27 06:27:59 -07:00
test_init_fallback_on_exhausted_pool.py	fix(agent): try fallback providers at init when primary credential pool is exhausted (#17929)	2026-05-02 02:09:46 -07:00
test_interactive_interrupt.py
test_interrupt_propagation.py	test: stop testing mutable data — convert change-detectors to invariants (#13363)	2026-04-20 23:20:33 -07:00
test_invalid_context_length_warning.py	fix(tests): resolve CI test failures — pool auto-seeding, stale assertions, mock isolation	2026-04-15 22:05:21 -07:00
test_iteration_budget_race.py	fix(run_agent): acquire lock in IterationBudget.used property	2026-05-04 12:37:28 -07:00
test_jsondecodeerror_retryable.py	fix(agent): retry on json.JSONDecodeError instead of treating it as a local validation error (#15107)	2026-04-24 05:02:58 -07:00
test_last_reasoning_per_turn.py	test: pin per-turn reasoning extraction semantics	2026-05-05 05:00:05 -07:00
test_long_context_tier_429.py
test_memory_nudge_counter_hydration.py	fix(agent): hydrate memory-nudge counters from conversation_history (#22774)	2026-05-09 12:48:03 -07:00
test_memory_provider_init.py	fix(memory): keep Honcho provider opt-in	2026-04-18 22:50:55 -07:00
test_memory_sync_interrupted.py	feat(memory): notify providers on mid-process session_id rotation (#17409)	2026-04-29 04:57:22 -07:00
test_message_sequence_repair.py	fix(run_agent): break permanent empty-response loop from orphan tool-tail (#21385)	2026-05-07 08:35:10 -07:00
test_openai_client_lifecycle.py
test_percentage_clamp.py	fix: update 6 test files broken by dead code removal	2026-04-10 03:44:43 -07:00
test_plugin_context_engine_init.py	fix(tests): make AIAgent constructor calls self-contained (#11755)	2026-04-17 12:32:03 -07:00
test_primary_runtime_restore.py	fix(agent): only set rate-limit cooldown when leaving primary; add tests	2026-04-24 05:35:43 -07:00
test_provider_attribution_headers.py	refactor(gmi): move User-Agent to profile.default_headers	2026-05-08 03:22:11 -07:00
test_provider_fallback.py	fix(fallback): skip chain entries matching current provider/model/base_url (#22780)	2026-05-09 12:48:19 -07:00
test_provider_parity.py	fix(aux): remove hardcoded Codex fallback model, drop Codex from auto chain (#17765)	2026-04-29 23:23:50 -07:00
test_real_interrupt_subagent.py	fix(tests): fix 78 CI test failures and remove dead test (#9036)	2026-04-13 10:50:24 -07:00
test_redirect_stdout_issue.py
test_repair_tool_call_arguments.py	fix(run_agent): handle unescaped control chars in tool_call arguments (#15356)	2026-04-24 15:06:41 -07:00
test_repair_tool_call_name.py	fix(agent): repair CamelCase + _tool suffix tool-call emissions (#15124)	2026-04-24 05:32:08 -07:00
test_review_prompt_class_first.py	fix(review): tell background reviewer not to capture transient env failures as skills (#23004)	2026-05-09 22:51:25 -07:00
test_run_agent.py	fix(agent): extract thinking from content-list blocks for DeepSeek V4 Pro	2026-05-09 13:36:12 -07:00
test_run_agent_codex_responses.py	fix(memory): drop scrub from interim commentary + final response	2026-04-27 12:37:33 -07:00
test_run_agent_multimodal_prologue.py	refactor: unify transport dispatch + collapse normalize shims	2026-04-22 18:34:25 -07:00
test_sequential_chats_live.py	test: regression guards for the keepalive/transport bug class (#10933) (#11266)	2026-04-16 16:36:33 -07:00
test_session_meta_filtering.py
test_session_reset_fix.py
test_steer.py	refactor(steer): simplify injection marker to 'User guidance:' prefix (#13340)	2026-04-20 22:18:49 -07:00
test_stream_drop_logging.py	feat(stream-retry): add upstream + timing diagnostics to drop log (#23005)	2026-05-09 22:49:35 -07:00
test_stream_interrupt_retry.py	fix: /stop now immediately aborts streaming retry loop	2026-04-25 09:51:39 -07:00
test_streaming.py	fix(copilot-acp): disable streaming path for CopilotACPClient	2026-04-28 11:33:07 -07:00
test_streaming_tool_call_repair.py	fix: repair malformed tool call args in streaming assembly before flagging as truncated	2026-04-24 15:03:07 -07:00
test_strict_api_validation.py
test_strip_reasoning_tags_cli.py	fix(display): strip standalone tool-call XML tags from visible text	2026-04-22 18:12:42 -07:00
test_switch_model_context.py	fix: pass config_context_length to switch_model context compressor	2026-04-10 05:52:45 -07:00
test_switch_model_fallback_prune.py	fix(agent): default missing fallback chain on switch	2026-04-24 05:35:43 -07:00
test_thinking_only_sanitizer.py	fix(agent): drop thinking-only assistant turns before provider call (#16959)	2026-04-28 03:50:51 -07:00
test_token_persistence_non_cli.py	fix: make session search initialize session db	2026-05-09 14:36:58 -07:00
test_tool_arg_coercion.py	fix(tools): wrap bare scalars in single-element list for array-typed args	2026-05-04 05:00:37 -07:00
test_tool_call_args_sanitizer.py	fix(agent): include name field on every role:tool message for Gemini compatibility (#16478)	2026-05-04 05:06:33 -07:00
test_tool_call_guardrail_runtime.py	fix(agent): make tool loop guardrails warning-first	2026-04-30 20:43:15 -07:00
test_tool_executor_contextvar_propagation.py	fix(agent): propagate ContextVars to concurrent tool worker threads (#18123)	2026-04-30 16:26:26 -07:00
test_unicode_ascii_codec.py	fix: always retry on ASCII codec UnicodeEncodeError — don't gate on per-component sanitization	2026-04-15 15:03:28 -07:00
test_vision_aware_preprocessing.py	feat(image-input): native multimodal routing based on model vision capability (#16506)	2026-04-27 06:27:59 -07:00