all publishers

zjunlp

@zjunlp source repo

345 published skills · page 3 of 4

  1. ▌
    Steering Coefficient Tuning · zjunlp
    How to set the strength of any additive intervention on internal representations — steering vectors, CAA, DAS dose-response, representation engineering, SAE feature scaling, ROME-style edits. Use whenever the plan pins a steering strength (`α`, `β`, `dose`, `magnitude`, `scale`, `coefficient`, `k`) to a fixed value or a narrow range, especially one copied from a paper. Covers the coarse-to-fine sweep strategy (start wide — e.g. `[1, 2, 4, 8, 16, 32, …]` or a few multiples of the current activation — then narrow in on the optimum), why the best coefficient is layer-dependent, the mid- vs late-layer behavior (late layers break into repetition / format-spam), when to stop raising the coefficient and switch feature / layer / method instead, and the requirement to score every sweep point on a target metric alongside a fluency / general-ability metric. Prevents two symmetric failures: TOO SMALL → effect drowned in noise → false "no causal effect"; TOO LARGE → off-distribution collapse → fluency breaks / a random di
    9 installs
  2. ▌
    Dissecting Factual Predictions · zjunlp bundle
    Analyze and dissect factual recall in auto-regressive language models using attention knockout, hidden state analysis, and intervention techniques on GPT-2 and GPT-J models
    9 installs
  3. ▌
    Rome Model Editing · zjunlp bundle
    Use this skill when you need to edit factual knowledge in large language models like GPT-2 or GPT-J, perform causal tracing to understand model behavior, or implement Rank-One Model Editing (ROME) to modify specific factual associations without retraining
    5 installs
  4. ▌
    Feature Dictionary Learning · zjunlp bundle
    Feature Dictionary Learning methods address the polysemanticity of neuron-level units by decomposing a dense internal activation (e.g. a residual-stream state or MLP output) into a sparse weighted sum of directions drawn from a large over-complete dictionary. The dictionary contains far more directions than the activation's original dimensionality, and the decomposition is constrained to use only a small number of them at once. Each direction in the dictionary plays the role of an interpretable "feature": its weight measures how strongly that feature is present, turning a black-box vector into a small set of human-readable components.
    9 installs
  5. ▌
    Finetune Hyperparameter Sweep · zjunlp
    Fine-tuning LR protocol — full FT, LoRA / QLoRA / DoRA / PEFT adapter, across SFT, DPO, and GRPO / PPO / RL objectives. Fires on ANY fine-tune, including (especially) when the plan already fixes a learning rate or copies one from a reference paper: a fixed LR is one grid point, never the answer, so LR is always swept over a wide range. The only test of an LR is whether the run hits the pass criteria in `task.md` or the milestone's own claim criteria — loss, grad norm, and reward curves are never acceptance evidence, no matter how good they look. Criteria not met = change LR and re-run. Triggers: `learning_rate`, `SFTTrainer`, `DPOTrainer`, `GRPOTrainer`, `LoraConfig`, `peft_config`, `full fine-tune`, `SFT`, `DPO`, `GRPO`, `RLHF`, `QLoRA`, `DoRA`, `PEFT`, `adapter`.
    7 installs
  6. ▌
    Linguistic Regions LLM · zjunlp bundle
    Use this skill when working with linguistic region analysis in Large Language Models, including data preprocessing for multilingual training, region-based model training with DeepSpeed, and extracting/visualizing linguistic regions in transformer models
    3 installs
  7. ▌
    Residual Stream States · zjunlp bundle
    Use this skill when working with LLM hallucination detection, probing internal representations of language models, analyzing model correctness, or conducting intrinsic evaluation experiments on models like Mistral and Llama-3
    5 installs
  8. ▌
    Saelens · zjunlp bundle
    Use this skill when working with Sparse Autoencoders (SAEs) for mechanistic interpretability of language models, including training SAEs, loading pre-trained SAEs, analyzing neural network features, or integrating SAEs with TransformerLens, HuggingFace Transformers, or other PyTorch-based models.
    9 installs
  9. ▌
    Ica Lens · zjunlp bundle
    Use this skill when applying Independent Component Analysis as a training-free interpretability lens — decomposing a target activation site (residual stream, MLP output, attention-head output, or any cached hook point) into maximally non-Gaussian directions and treating each direction as a candidate monosemantic component for sparse-probing, targeted perturbation, annotation, and cross-comparison against trained SAE / transcoder / crosscoder features.
    10 installs
  10. ▌
    Static Parameters · zjunlp bundle
    Analyze and manipulate massive values in LLM attention mechanisms, particularly for understanding contextual knowledge processing in transformer models with RoPE
    12 installs
  11. ▌
    Sae Feature Activation State · zjunlp bundle
    Use this skill when working with Sparse Autoencoders (SAEs) for feature analysis, particularly for studying feature splitting, absorption, and attribution in language models. Activate for tasks involving SAE feature ablation, probing experiments, or analyzing how SAE latents affect model outputs in spelling and token-level tasks.
    10 installs
  12. ▌
    Dynamic Components · zjunlp bundle
    Identify and manipulate language-specific neurons in multilingual Large Language Models (LLMs) to understand and control language-specific behaviors in models like LLaMA-2, BLOOM, OPT, Mistral, and Phi-2
    9 installs
  13. ▌
    Representation And Parameter Analysis · zjunlp bundle
    Representation and Parameter Analysis interprets and controls a model by directly manipulating its two kinds of internal objects — features (the hidden-state activations produced during a forward pass) and weights (the parameters of the target model). The analysis side relies on standard linear-algebra tools applied straight to these objects: techniques such as PCA, mean-difference, or simple linear classifiers are run over collected features (e.g., from contrastive prompt pairs) or over weight differences between a fine-tuned and a pre-trained checkpoint, in order to extract a small set of meaningful directions. The control side then operates directly through arithmetic on those objects: adding such a direction back into the features steers behavior at inference time, while adding, subtracting, or combining task-level weight differences edits the model in one shot — so that interpretability findings translate into behavioral control without any retraining.
    9 installs
  14. ▌
    Crosscoder Learning · zjunlp bundle
    Use this skill when working with sparse autoencoders (SAEs), crosscoders, dictionary learning on neural network activations, training SAEs/crosscoders from scratch, loading pretrained dictionaries, caching model activations, or comparing model internals across fine-tuned model pairs using the dictionary_learning / crosscoder_learning library.
    10 installs
  15. ▌
    Transcoder Circuits · zjunlp bundle
    Use this skill when working with transcoder-based circuit analysis of large language models, including training transcoders, analyzing MLP sublayers, reverse-engineering LLM circuits, and creating feature dashboards for interpretability research.
    9 installs
  16. ▌
    Attribution Patching · zjunlp bundle
    Use when analyzing neural network circuits, performing attribution patching, automated circuit discovery, or investigating model interpretability through edge attribution methods in transformer models
    11 installs
  17. ▌
    Intermediate Outputs · zjunlp bundle
    Use this skill when working with circuit discovery in language models, mechanistic interpretability, activation patching, attribution patching, or Layer-wise Relevance Propagation (LRP) for neural network analysis
    6 installs
  18. ▌
    Neuron Value Weight · zjunlp bundle
    Analyze transformer feed-forward layers as key-value memories, extract activations, identify trigger examples, and compute key-value agreement in transformer language models
    9 installs
  19. ▌
    Attention Head Output · zjunlp bundle
    Use this skill when you need to analyze and interpret the internal workings of large language models layer by layer, visualize hidden states and predictions across transformer layers, or understand how models like Llama-3.1-8B and Qwen-2.5-7B make predictions at each layer using the Logit Lens technique
    5 installs
  20. ▌
    Residual Stream State · zjunlp bundle
    Use this skill when working with transformer model interpretability, analyzing layer-by-layer predictions, training tuned lenses to understand intermediate representations, or peeking into iterative computations of transformers
    7 installs
  21. ▌
    Layer Wise Representation · zjunlp bundle
    Use this skill when you need to enhance the truthfulness of Large Language Models (LLMs) or reduce hallucinations in model outputs. This skill provides TruthX, an inference-time method that edits LLM internal representations to control truthfulness and mitigate hallucinations.
    9 installs
  22. ▌
    Convrfm · zjunlp bundle
    Use this skill when working with convolutional neural feature ansatz (CNFA), deep convolutional recursive feature machines (Deep ConvRFM), or tasks involving kernel-based feature learning with convolutional architectures, VGG networks, patch-based Jacobian computation, and EGOP (expected gradient outer product) analysis on image datasets such as ImageNet and CIFAR.
    5 installs
  23. ▌
    Recursive Feature Machines · zjunlp bundle
    Use this skill when working with Recursive Feature Machines (RFM) for kernel-based machine learning, feature learning, or when implementing backpropagation-free models that learn features similarly to neural networks.
    8 installs
  24. ▌
    Inputs And Layer Wise States · zjunlp bundle
    Analyze and visualize layer-wise gradient behaviors in LLMs during fine-tuning for fast vs slow thinking tasks, calculate gradient statistics, and understand training patterns across different model layers
    7 installs
  25. ▌
    Eap Ig · zjunlp bundle
    Use this skill to discover and analyze computational circuits within autoregressive transformer language models via attribution methods like Edge Attribution Patching with Integrated Gradients (EAP-IG), enabling mechanistic interpretability studies and circuit evaluation.
    5 installs
  26. ▌
    Automatic Circuit Discovery · zjunlp bundle
    Activates when working on mechanistic interpretability of transformers, specifically for automated circuit discovery in transformer models using computational graph editing and analysis.
    9 installs
  27. ▌
    Tp4 Feature Learning · zjunlp bundle
    Use this skill when working with infinite-width neural networks for feature learning, replicating Word2Vec or MAML experiments from the Tensor Programs series (TP4), or implementing infinite-width limits (GP, NTK, muP) for meta-learning and word embedding tasks.
    8 installs
  28. ▌
    Steer Features · zjunlp bundle
    Use this skill for feature-level steering of models — locating the internal feature that drives a target behavior, scoring and selecting it by its effect on the model's output, and directly amplifying or shrinking that feature's activation during generation to control behavior. Applies to features read from the model's own activations or from a Sparse Autoencoder (SAE); the bundled demo scripts happen to use an SAE, but the method does not require one.
    7 installs
  29. ▌
    Deep Neural Feature Ansatz · zjunlp bundle
    Use this skill when working with the Deep Neural Feature Ansatz (DNFA) — verifying feature learning in neural networks, training fully connected networks on image/tabular datasets, computing Neural Tangent Kernels (NTK) and Neural Network Gaussian Processes (NNGP), or reproducing experiments from the paper "The Deep Neural Feature Ansatz" (arXiv:2212.13881).
    11 installs
  30. ▌
    Steering Vectors · zjunlp bundle
    Work with Llama 2 model steering using Contrastive Activation Addition (CAA) for behavioral control, including generating steering vectors, evaluating model behavior, and analyzing activation patterns
    8 installs
  31. ▌
    Representation Engineering · zjunlp bundle
    Use this skill when working with Representation Engineering (RepE) for AI transparency, monitoring, or controlling internal representations of large language models including truthfulness detection, emotion control, harmlessness steering, and memorization analysis.
    9 installs
  32. ▌
    Task Vectors · zjunlp bundle
    Use this skill when working with task arithmetic for editing neural network models, including creating task vectors from pre-trained and fine-tuned checkpoints, combining them via arithmetic operations (negation, addition, analogies), and applying them to CLIP vision models for multi-task learning.
    9 installs
  33. ▌
    CSV Analysis · zjunlp
    Use this skill for CSV data analysis tasks that require reading a local CSV file, checking row counts and columns, grouping records, computing rates or aggregates, creating a chart, and writing a short Markdown report.
    4 installs
  34. ▌
    Applicable Fee Ids · zjunlp
    Solve questions about which fee IDs apply to a payment merchant, transaction characteristics, or time period in the dabstep dataset. Use this skill for any question asking "which fee IDs apply to X", "what are the applicable fee IDs for merchant Y", "which merchants are affected by fee Z", or any query involving filtering fees.json based on payment characteristics or merchant attributes.
    6 installs
  35. ▌
    Average Fee Estimation · zjunlp
    Solve dabstep Average_Fee_Estimation problems. Use this skill for questions asking about average payment processing fees, which card scheme is cheapest/most expensive in an average scenario, and fee calculations based on card scheme, account type, MCC description, credit/debit type, or any combination of these filters.
    6 installs
  36. ▌
    Longds Bench · zjunlp bundle
    Self-evaluate the current agent on LongDS-Bench (zjunlp/DataMind): the long-horizon, multi-turn agentic data-analysis benchmark. Use this when the user asks to run, score, or benchmark an agent on LongDS / LongDS-Bench / DataMind longds, or to measure multi-turn data-analysis ability. This does NOT use DSGym's Docker runtime — the agent running this skill IS the agent under test: it reads a locally-prepared dataset, performs the multi-turn analysis with its own tools, and is scored by the official LLM-judge rule. The ~19.5 GB dataset must be downloaded and prepared by the operator beforehand (see `$SKILL_DIR/../README`); this skill does not download it. Heavyweight and long-running; run in the background if supported and confirm scope first.
    11 installs
  37. ▌
    Globem User Analysis · zjunlp
    Comprehensive individual-user analysis on the GLOBEM dataset — a longitudinal passive-sensing + mental-health study of college students. Use this skill whenever a task involves analyzing a specific participant (e.g. "Analyze user INS-W_002") from the GLOBEM dataset, exploring behavioral patterns from smartphone sensors, correlating behavioral signals with mental health outcomes, or producing a comprehensive user profile from multimodal sensing data.
    5 installs
  38. ▌
    Mimic Patient Analysis · zjunlp
    Comprehensive patient analysis using the MIMIC-IV clinical database. Use this skill whenever asked to analyze, summarize, or investigate a patient's medical history, hospital admissions, diagnoses, medications, procedures, or clinical course from a MIMIC-IV SQLite database. Triggers on prompts like "Analyze patient [ID]", "summarize patient history", "what happened to patient X", or any request to explore patient-level EHR data from MIMIC-IV tables.
    5 installs
  39. ▌
    Total Fees Calculation · zjunlp
    Skill for computing total payment processing fees for a merchant over a specific day, date range, or month in the dabstep dataset. Use this skill whenever the question asks for "total fees", "fees paid", or "fees charged" for a merchant over some time period. The computation requires matching each transaction to a fee rule from fees.json using multiple merchant and transaction properties, then summing fixed_amount + rate * eur_amount / 10000 for each matched transaction.
    12 installs
  40. ▌
    Sec 10k Company Analysis · zjunlp
    Comprehensive analysis of public companies using SEC EDGAR 10-K financial data stored in a SQLite database. Use this skill whenever the task involves analyzing a company by CIK (Central Index Key), querying SEC financial data, exploring a financial database with tables like companies, filings, financial_facts, or producing a structured financial analysis report. Covers schema navigation, metric discovery, industry-specific exploration, multi-year trend analysis, income statement structure, capital returns analysis, long-term obligations, and producing an insightful financial summary that connects metrics across dimensions.
    10 installs
  41. ▌
    Routing And Cost Optimization · zjunlp
    Solves payment routing and cost optimization problems in the dabstep dataset. Use when the question asks which card scheme a merchant should steer traffic to (for minimum or maximum fees), or which ACI (Authorization Characteristics Indicator) to incentivize for fraudulent transactions to achieve the lowest possible fees. Triggers on: "steer traffic", "card scheme", "minimum fees", "maximum fees", "ACI", "incentivize", "fraudulent transactions", "lowest possible fees", payment routing.
    10 installs
  42. ▌
    Average Transaction Value Stats · zjunlp
    Skill for computing average transaction value statistics from payment transaction data in the dabstep dataset. Use this skill when a question asks about average transaction amount/value grouped by a categorical field (e.g., shopper_interaction, issuing_country, acquirer_country, aci), possibly filtered by merchant, card scheme, and/or date range.
    4 installs
  43. ▌
    Fee Delta And Impact Simulation · zjunlp
    Fee Delta and Impact Simulation
    4 installs
  44. ▌
    Fraud And General Macro Analysis · zjunlp
    Solve fraud analysis and general macro transaction analysis questions on the dabstep payment dataset. Use this skill for questions about fraud rates, transaction distributions, merchant rankings, card scheme analysis, country breakdowns, correlation studies, and yes/no comparative questions on payment data.
    6 installs
  45. ▌
    Dataset Metadata And Business Rules · zjunlp
    Answers questions about the structure, metadata, field definitions, and business rules of the dabstep payment processing dataset. Use this skill when questions ask about: column names, field meanings, fee rule structure, which factors affect fees, fee formula, boolean factor effects on cost, volume/fraud/capture-delay tier meanings, or whether specific fee types exist. Also use for "Not Applicable" determination when a business concept is mentioned but no corresponding fee or data field exists.
    7 installs
  46. ▌
    Dabstep Dataset Metadata Business Rules · zjunlp
    Solve Dataset_Metadata_and_Business_Rules questions in the dabstep payment processing dataset. Use this skill for questions about dataset column meanings, fee factor directionality, business rules encoded in fee structures, "Not Applicable" detection, and metadata interpretation for the dabstep payment processing domain. Trigger whenever questions reference payment fees, fraud indicators, boolean fee factors, volume tiers, ACI codes, or any field from the dabstep dataset.
    8 installs
  47. ▌
    Highest Cost Scenario Identification · zjunlp
    Solve dabstep dataset questions that ask to identify the most expensive Merchant Category Code (MCC) or the most expensive Authorization Characteristics Indicator (ACI) for a given transaction. Use this skill when the question asks "what is the most expensive MCC for a transaction of X euros" or "what is the most expensive ACI for a credit transaction of X euros on [CardScheme]".
    5 installs
  48. ▌
    Ddr Globem Analysis · zjunlp
    Analyze a specific participant's longitudinal passive-sensing and psychological data in the GLOBEM digital depression research dataset. Use this skill whenever the task involves: analyzing a user's mental health or behavioral data from wearables/smartphones, generating QA pairs about behavioral/psychological changes over time, working with EMA, depression scores, activity, sleep, communication, location, or phone-usage data, or any user-profile analysis in the DDR/GLOBEM context.
    9 installs
  49. ▌
    Mimic Iv Patient Analysis · zjunlp
    Comprehensive strategy for analyzing individual patient records in MIMIC-IV EHR database and generating high-quality, diverse QA pairs. Use this skill whenever the task involves analyzing a specific patient's clinical data from MIMIC-IV (or similar EHR databases), querying across hospital and ICU tables, and submitting QA pairs that cover the patient's full clinical story — diagnoses, procedures, medications, care trajectory, and outcomes. Trigger when you see tasks like "Analyze patient <ID>", "Generate QA pairs for patient", or any patient-centric EHR exploration task.
    11 installs
  50. ▌
    Webshop Query Interpreter · zjunlp bundle
    Interprets a user's shopping query to extract key product requirements such as item type, attributes, and constraints. Use when a new shopping instruction is received and you need to parse natural language into structured criteria (product category, desired features, price limits) before performing a search. The output is a clear set of search parameters for use in subsequent product lookup steps.
    7 installs
  51. ▌
    Alfworld Goal Interpreter · zjunlp bundle
    Parses the natural language task goal to extract actionable sub-objectives and required objects. Trigger this skill whenever a new task is assigned to break down complex instructions into clear, sequential targets. It interprets phrases like 'look at X under Y' to identify target objects (pillow), reference objects (desklamp), and spatial relationships (under).
    11 installs
  52. ▌
    Alfworld Storage Explorer · zjunlp bundle
    Systematically explores storage receptacles (drawers, cabinets, shelves) to find an appropriate placement location for an object. Use when the agent needs to store an item but the exact target receptacle is unknown or ambiguous. Opens, inspects, and closes candidate receptacles to assess suitability, then places the object in the best match.
    6 installs
  53. ▌
    Task Completion Focus · zjunlp bundle
    Focuses on a specific target object to signal task completion. Use when you have produced the required final object (like a grown banana) and need to formally complete the assigned task. This handles the 'focus on OBJ' action that typically marks successful task execution in the environment.
    8 installs
  54. ▌
    Webshop Attribute Selector · zjunlp
    Selects specific product attributes (e.g., flavor, size) on a product detail page to match user requirements. Triggered when a product page presents configurable options and the parsed query specifies needed values. It identifies the correct option from available clickables (e.g., 'smoked bacon' for flavor, '4 ounce (pack of 2)' for size) and executes the selection click.
    8 installs
  55. ▌
    Webshop Attribute Verifier · zjunlp
    Verifies product attributes on a web shop detail page by extracting price, comparing color availability, validating specifications, and confirming option selections against user requirements before purchase. Use when you need to check if a product matches requirements, verify product details before buying, confirm item specifications on an online store product page, or validate that price, color, size, or other attributes satisfy the user's constraints. Outputs a `Thought:` assessment followed by a `click[value]` action to select the matching option and proceed, or navigates back to search if the product does not match.
    11 installs
  56. ▌
    Webshop Purchase Initiator · zjunlp bundle
    Initiates the purchase process by clicking the buy button once a product and its options have been selected and verified against requirements. Use when you are on a product detail page, all selection criteria (price, features, options) are satisfied, and the user is ready to proceed to checkout. Confirms the final price meets budget constraints before executing click[buy now].
    9 installs
  57. ▌
    Webshop Search Constructor · zjunlp bundle
    This skill formulates an effective search query from structured product criteria. It is used when initiating a product search on an e-commerce platform, converting attributes into a concise, relevant keyword string. The input is a set of parsed criteria (e.g., 'jewelry box', '10 slots', 'easy to clean', 'price < $60'), and the output is a formatted search action string (e.g., 'search[jewelry box, 10 slots, less than $60, easy to clean]') optimized for the platform's search engine.
    5 installs
  58. ▌
    Alfworld Receptacle Closer · zjunlp bundle
    Closes an open receptacle to maintain environment tidiness after inspection. Use when you have finished searching a container (drawer, cabinet, fridge) and no longer need it open. Takes a receptacle identifier as input and outputs confirmation that the receptacle is closed, preventing clutter during multi-step search tasks.
    4 installs
  59. ▌
    Alfworld Receptacle Finder · zjunlp bundle
    Searches for a suitable empty or appropriately occupied receptacle (like a shelf or table) to place an object. Use when you are holding an object that needs to be stored or placed and must find a receptacle that meets the placement criteria. Examines candidate receptacles by navigating to and inspecting each one until a suitable location is found.
    11 installs
  60. ▌
    Alfworld Receptacle Opener · zjunlp
    This skill opens a closed receptacle to allow access to its contents or to enable interaction with it. It should be triggered when the agent encounters a closed receptacle (e.g., 'The fridge 1 is closed.') that must be opened to proceed with a task, such as cooling an item with a fridge or taking an object from inside a cabinet. The skill outputs the 'open' action on the specified receptacle.
    10 installs
  61. ▌
    Scienceworld Tool User · zjunlp bundle
    Uses a tool from inventory on a target object or location to perform a specific environmental interaction, such as digging, cutting, or measuring. Use this skill when a task requires modifying the environment or manipulating materials with a tool (e.g., using a shovel to dig soil, a thermometer to measure temperature, or an axe to cut wood). Takes the tool and target as inputs and outputs the result of the interaction.
    10 installs
  62. ▌
    Alfworld Appliance Preparer · zjunlp bundle
    Prepares a household appliance (microwave, oven, toaster, fridge) for use by ensuring it is in the correct open/closed state. Use when the agent needs to heat, cool, or cook an item and must first open or close the appliance before placing an object inside. Takes an appliance identifier as input and outputs a confirmation that the appliance is ready for the next action.
    10 installs
  63. ▌
    Alfworld Navigation Planner · zjunlp bundle
    Plans a path to move the agent between receptacles to search for target objects. Use this when you need to traverse the environment to reach a specific location or systematically explore multiple areas. It takes the current location and destination receptacle as input, and outputs the next 'go to' action to approach the target.
    5 installs
  64. ▌
    Alfworld Object Transporter · zjunlp bundle
    Picks up a target object from its current receptacle and moves it to a specified destination receptacle. Use when you have located an object and need to relocate it to complete a task (e.g., moving a laptop to a desk). Takes the object identifier, source receptacle, and destination receptacle as inputs and outputs the action sequence to take, transport, and place the object.
    12 installs
  65. ▌
    Webshop Product Detail Check · zjunlp bundle
    Examines a specific product's detailed page to verify it matches the user's requirements, checking price, description, features, and reviews. Use when you have navigated to a product detail page from search results and need to confirm the product meets all user-specified constraints before purchasing. It provides a final suitability assessment with a clear proceed-or-reject recommendation.
    5 installs
  66. ▌
    Alfworld Appliance Navigator · zjunlp
    Navigates the agent to a target appliance (microwave, stove, fridge, or sinkbasin) needed for object processing. Use when you are holding an object that needs heating, cooling, or cleaning and must move to the correct appliance station. Identifies the required appliance from the task context and executes the movement action.
    7 installs
  67. ▌
    Alfworld Environment Scanner · zjunlp bundle
    Performs an initial scan of the ALFWorld environment to identify all visible objects and receptacles. Use when you first enter an environment and need to build a mental map for task planning. Processes raw observation text into a structured list of entities, categorizing them as objects or receptacles.
    11 installs
  68. ▌
    Alfworld Receptacle Searcher · zjunlp bundle
    This skill searches for a specific receptacle (e.g., garbage can, cabinet) by systematically exploring the environment, checking multiple locations until found. Trigger when the target receptacle is not initially visible, often after scanning nearby areas. It outputs the receptacle's location and may involve navigating through multiple waypoints.
    8 installs
  69. ▌
    Scienceworld Task Parser · zjunlp bundle
    Analyzes user instructions in ScienceWorld environments to extract specific task requirements and constraints. Use when receiving a new task to identify required objects, target locations, and action sequences before taking any environment actions.
    14 installs
  70. ▌
    Alfworld Clean Object · zjunlp bundle
    Cleans a specified object using an appropriate cleaning receptacle (e.g., sinkbasin). Use when a task requires an object to be in a clean state (e.g., "clean potato", "wash apple") before proceeding. Navigates to the cleaning location, performs the clean action, and confirms the object is now clean.
    10 installs
  71. ▌
    Alfworld Tool Locator · zjunlp bundle
    Searches for a specified tool or device (e.g., a desklamp, knife, or sponge) within the ALFWorld environment by checking relevant surfaces. Use when you need a tool to interact with another object as part of a task but the tool is not in your inventory or immediate vicinity. Takes a tool name as implicit input and navigates to likely storage spots (sidetables, shelves, countertops) until the tool is found.
    6 installs
  72. ▌
    Webshop Initial Search · zjunlp
    Performs the first search on an e-commerce platform using keywords derived from parsed user requirements. Use when starting product discovery from a user instruction or when previous search results are insufficient and a new query is needed. It formulates a search query from the criteria (e.g., '24 pack of 7.5 ounce bottles of non-gmo classic tonic') and executes the search[] action, returning the initial result page.
    6 installs
  73. ▌
    Webshop Product Search · zjunlp
    Performs an initial product search using a web interface by generating appropriate search keywords based on interpreted query criteria. Use when starting a product discovery task, when returning to search results after rejecting a product, or when the observation contains a search interface prompt. The skill inputs structured search parameters and outputs a list of candidate products from the search results page.
    5 installs
  74. ▌
    Alfworld Object Cooler · zjunlp bundle
    Cools a held object using an appropriate cooling appliance such as a fridge or freezer. Use when the task requires reducing the temperature of an object (e.g., "cool some pot", "chill the mug") and the agent is already holding the object. Performs the ALFWorld `cool` action and outputs the cooled object ready for subsequent placement or serving steps.
    13 installs
  75. ▌
    Alfworld Object Heater · zjunlp bundle
    Heats a specified object using an available heating appliance (e.g., microwave, stoveburner). Use when you are holding an object that requires heating and need to navigate to and operate the heating appliance. Takes the object and appliance as inputs and results in the object being in a heated state.
    11 installs
  76. ▌
    Alfworld Object Placer · zjunlp bundle
    Places a held object onto or into a target receptacle. Trigger this skill when the agent is carrying an object and needs to deposit it at a specific location to complete a task, such as installing an item on a holder or storing it in a container. It takes the object and destination receptacle as inputs, performing a 'put' action to finalize the object's positioning according to the task goal.
    9 installs
  77. ▌
    Alfworld Object Storer · zjunlp bundle
    Use when the agent is holding an object and needs to place it into a target receptacle in ALFWorld. This skill checks receptacle suitability, opens closed containers if needed, and executes the `put` command to store the object. It handles both open surfaces (countertops, beds) and closed containers (drawers, cabinets).
    5 installs
  78. ▌
    Alfworld Task Verifier · zjunlp bundle
    Use when the agent needs to check whether an ALFWorld task objective has been met after completing a sub-action (e.g., placing an object). This skill parses the task goal, evaluates the latest environment observation, and outputs a verification decision — task complete, task incomplete, or action ineffective — to guide the next step.
    10 installs
  79. ▌
    Webshop Result Analyzer · zjunlp bundle
    This skill evaluates a list of search results against the target product criteria. Trigger it when a search result page is observed, to identify promising product listings. It analyzes each result's title, price, and brief description to shortlist items that best match the required attributes (like size, color, and price), outputting a prioritized list of candidate product IDs or links for further inspection.
    12 installs
  80. ▌
    Webshop Search Executor · zjunlp bundle
    Executes a search on an e-commerce platform using parsed keywords. Use when you have extracted search terms from a user query and need to perform the actual search[] action on the WebShop interface. Takes structured search terms and performs a search action, returning a list of product results for evaluation.
    5 installs
  81. ▌
    Webshop Variant Chooser · zjunlp bundle
    This skill selects a product variant (e.g., size, quantity) from available options on a product detail page. It is triggered when a product page displays multiple choices, assessing the options (like '1.1 pound (pack of 1)' or 'pack of 2') and selecting one that aligns with user needs, often defaulting to a standard single unit if unspecified.
    8 installs
  82. ▌
    Alfworld Object Locator · zjunlp bundle
    Use when the agent needs to find a specific object in ALFWorld that is not currently in inventory and whose location is unknown. This skill parses the environment observation, ranks receptacles by likelihood of containing the target object using common-sense reasoning, and outputs a navigation action to the most promising location.
    9 installs
  83. ▌
    Webshop Product Selector · zjunlp bundle
    Selects the most promising product candidate from a filtered shortlist for detailed inspection. Use when you are on a search results page with multiple viable items and need to make a final selection based on a heuristic such as best feature match or lowest price. Triggers navigation to the selected product's detail page via click[product_id].
    9 installs
  84. ▌
    Alfworld Device Operator · zjunlp bundle
    Operates a device or appliance (like a desklamp, microwave, or fridge) to interact with another object. Use when the task requires using a tool on a target item (e.g., "look at laptop under the desklamp", "heat potato with microwave"). Locates both the device and target object, co-locates them, and executes the appropriate use action (toggle, heat, cool, or clean).
    8 installs
  85. ▌
    Alfworld Search Verifier · zjunlp bundle
    Re-examines previously visited locations to confirm the absence of a target object or to check for overlooked items. Use when an initial search fails to find enough objects or when double-checking is required before concluding task failure. Systematically revisits receptacles, re-opens closed containers, and re-inspects contents to ensure no viable location was missed.
    4 installs
  86. ▌
    Webshop Product Evaluator · zjunlp bundle
    Evaluates product listings against user requirements such as price limits and feature matches to identify viable options. Use when you are on a search results page containing multiple products and need to select the most promising candidate for detailed inspection. The skill analyzes product titles, prices, and brief descriptions to rank and choose the best match.
    6 installs
  87. ▌
    Webshop Purchase Executor · zjunlp bundle
    Executes the purchase action for a confirmed suitable product on an e-commerce page. Use when a product has been verified against the user's requirements (price, attributes) and the decision is to buy. Performs a click[Buy Now] action on the purchase button to complete the transaction.
    10 installs
  88. ▌
    Skillnet · zjunlp bundle
    Search, download, create, evaluate, and analyze reusable agent skills via SkillNet — the open skill supply chain for AI agents. Use when: (1) Before any multi-step task — search SkillNet for existing skills first, (2) User says "find a skill", "learn this repo/doc", "turn this into a skill", or mentions skillnet, (3) User provides a GitHub URL, PDF, DOCX, PPT, execution logs, or trajectory — create a skill from it, (4) After completing a complex task with non-obvious solutions — create a skill to preserve learnings, (5) User wants to evaluate skill quality, organize/analyze a local skill library, or orchestrate a supported scene. NOT for: single trivial operations (rename variable, fix typo), or tasks with no reusable knowledge.
    11 installs
  89. ▌
    Alfworld Tool User · zjunlp bundle
    Use when the agent needs to apply a tool to a target object in ALFWorld to accomplish an interaction such as cleaning, heating, cooling, or examining. This skill handles locating both the tool and target object, then executing the correct environment action (e.g., `clean`, `heat`, `cool`, `use`) to progress the task.
    11 installs
  90. ▌
    Webshop Result Filter · zjunlp bundle
    Filters search results by evaluating product listings against specific user constraints like price, features, or ratings. Use when you are on a search results page and need to systematically identify which products meet all given criteria before selecting one for closer inspection. Takes a list of products with their details and outputs a filtered subset that meets the defined requirements.
    7 installs
  91. ▌
    Scienceworld Measurement Taker · zjunlp bundle
    Use when the agent needs to measure a quantitative property (temperature, weight, pH) of a target object or substance using a measurement tool. This skill covers acquiring the tool, preparing both tool and target with focus actions, executing the measurement via "use [TOOL] on [TARGET]", and interpreting the resulting value for conditional decisions such as classification or placement.
    6 installs
  92. ▌
    Scienceworld Object Classifier · zjunlp bundle
    Moves a tested or examined object into a designated container (e.g., a specific colored box) based on a determined property. Use when you have completed a test or inspection and need to fulfill a classification or sorting subtask. It takes the object and target container as inputs and performs the move action.
    10 installs
  93. ▌
    Scienceworld Substance Fetcher · zjunlp bundle
    Locates and retrieves a target substance or material from a container in the environment. Use this skill when the task requires processing a specific substance (e.g., chocolate, sodium chloride, a chemical) and you need to find and acquire it. Searches rooms and containers (fridge, cupboard, counter) and retrieves the substance via pick up or move actions.
    10 installs
  94. ▌
    Alfworld Heat Object With Appliance · zjunlp bundle
    Uses a heating appliance (microwave, stoveburner, oven) to apply heat to a specified object. Use when the task requires warming or cooking an item (e.g., "heat some egg", "warm the mug") and a heating appliance is available. Takes the object name and appliance name as input and outputs the object in a heated state, ready for placement at the task's target location.
    26 installs
  95. ▌
    Scienceworld Conditional Placer · zjunlp bundle
    Places an object into one of several designated containers based on a measured condition, such as a temperature threshold. Use this skill when you have completed a measurement or assessment and the task requires sorting or storing the object into one of multiple containers according to a rule (e.g., "if temperature > X, place in container A; otherwise container B").
    8 installs
  96. ▌
    Scienceworld Container Transfer · zjunlp bundle
    Moves a substance or object from one container to another (e.g., moving lead to a metal pot). Use this skill when you need to prepare materials for processing, such as transferring a substance into a heat-resistant container before heating, or reorganizing materials between vessels for mixing or measurement.
    4 installs
  97. ▌
    Controlled Waiting · zjunlp bundle
    Executes timed waiting using 'wait' or 'wait1' actions to advance the simulation clock. Use when a time-dependent process like plant growth, chemical reaction, or mechanical cycle must progress before you can continue the task.
    9 installs
  98. ▌
    Scienceworld Instruction Reader · zjunlp bundle
    Reads a recipe or note from the inventory using the 'read' action and extracts key information. Use this skill when you have acquired a recipe, note, or readable document in your inventory and need to extract explicit instructions, ingredient lists, or procedural steps before executing a task.
    11 installs
  99. ▌
    Scienceworld Liquid Source Finder · zjunlp bundle
    This skill searches rooms for a liquid source (e.g., sink, toilet, jug) by sequentially surveying different locations. It should be triggered when the task requires obtaining a liquid (like water) and the agent does not have immediate access to it. The skill involves teleporting to candidate rooms, looking around, and identifying potential sources, guiding the agent toward the nearest available liquid.
    9 installs
  100. ▌
    Scienceworld Container Inspector · zjunlp bundle
    Inspects the contents of a container or device using the 'look at' action. Use this skill when you need to verify what is inside a container (e.g., checking if lead is in the blast furnace), monitor the state of contents (e.g., solid vs. liquid phase changes), or confirm that a placement or process step succeeded. Returns a detailed list of contents and their states for process monitoring.
    7 installs