plug.directory

`SmolFactory__start_with_overrides` – 1 MCP server offers this tool

What it usually takes

adv_batch_size_v – in 1 of 1
adv_dataset_format_v – in 1 of 1
adv_dataset_name_v – in 1 of 1
adv_dataset_split_v – in 1 of 1
adv_developer_message_v – in 1 of 1
adv_eval_steps_v – in 1 of 1
adv_gas_v – in 1 of 1
adv_gpt_mode_v – in 1 of 1
adv_input_field_v – in 1 of 1
adv_logging_steps_v – in 1 of 1
adv_lora_alpha_v – in 1 of 1
adv_lora_dropout_v – in 1 of 1

No server marks it as only reading: treat it as able to change data.

From each publisher's own tool list. We never call a tool.

Servers

Server	Inputs	Read-only?
SmolFactory `co.huggingface.tonic/smolfactory`	`adv_batch_size_v`,`adv_dataset_format_v`,`adv_dataset_name_v`,`adv_dataset_split_v`,`adv_developer_message_v`,`adv_eval_steps_v`,`adv_gas_v`,`adv_gpt_mode_v`,`adv_input_field_v`,`adv_logging_steps_v`,`adv_lora_alpha_v`,`adv_lora_dropout_v`,`adv_lora_r_v`,`adv_lr_v`,`adv_max_grad_norm_v`,`adv_max_length_v`,`adv_max_samples_v`,`adv_max_seq_length_v`,`adv_med_batch_size_v`,`adv_med_dataset_config_v`,`adv_med_developer_message_v`,`adv_med_gradient_accumulation_steps_v`,`adv_med_learning_rate_v`,`adv_med_max_seq_length_v`,`adv_med_num_train_epochs_v`,`adv_med_system_message_v`,`adv_min_length_v`,`adv_min_lr_num_v`,`adv_mixed_precision_v`,`adv_model_identity_v`,`adv_num_train_epochs_v`,`adv_num_workers_v`,`adv_quantization_type_v`,`adv_save_steps_v`,`adv_sm_batch_size_v`,`adv_sm_dataset_name_v`,`adv_sm_eval_steps_v`,`adv_sm_filter_bad_entries_v`,`adv_sm_gas_v`,`adv_sm_input_field_v`,`adv_sm_lc_batch_size_v`,`adv_sm_lc_dataset_name_v`,`adv_sm_lc_eval_steps_v`,`adv_sm_lc_filter_bad_entries_v`,`adv_sm_lc_gas_v`,`adv_sm_lc_input_field_v`,`adv_sm_lc_learning_rate_v`,`adv_sm_lc_logging_steps_v`,`adv_sm_lc_max_iters_v`,`adv_sm_lc_max_seq_length_v`,`adv_sm_lc_model_name_v`,`adv_sm_lc_no_think_system_message_v`,`adv_sm_lc_sample_seed_v`,`adv_sm_lc_sample_size_v`,`adv_sm_lc_save_steps_v`,`adv_sm_lc_target_field_v`,`adv_sm_lc_use_chat_template_v`,`adv_sm_lc_warmup_steps_v`,`adv_sm_learning_rate_v`,`adv_sm_logging_steps_v`,`adv_sm_max_seq_length_v`,`adv_sm_mode_v`,`adv_sm_model_name_v`,`adv_sm_sample_seed_v`,`adv_sm_sample_size_v`,`adv_sm_save_steps_v`,`adv_sm_target_field_v`,`adv_system_message_v`,`adv_target_field_v`,`adv_warmup_ratio_v`,`adv_wd_v`,`advanced_enabled_v`,`author_name_v`,`config_choice_v`,`create_dataset_repo_v`,`deploy_trackio_space_v`,`experiment_name_v`,`min_lr_rate_v`,`min_lr_v`,`model_description_v`,`model_family_v`,`monitoring_mode_v`,`push_to_hub_v`,`repo_short_v`,`scheduler_override_v`,`switch_to_read_after_v`,`trackio_space_name_v`,`trainer_type_v`	Not stated