Skills
Automated Documentation Coverage and Accuracy Auditor
Identifies documentation gaps and inaccuracies by comparing the main branch implementation and configuration against the existing docs directory. It proposes targeted updates to ensure documentation remains synchronised with the codebase.
Automated Code Change Verification Pipeline
Executes a mandatory verification pipeline including formatting, linting, type checking, and testing to ensure code integrity. It uses fail-fast semantics and parallel execution for validation steps following code formatting.
Automate Grovs Mobile Project Setup
Automate the provisioning of Grovs instances and projects, including SDK configuration for iOS and Android and redirect behaviour setup. This skill handles the end-to-end orchestration of mobile deep linking and tracking parameters.
Grovs Tracking Link Management
Search, list, and update tracking link metadata, tags, and redirects using numeric IDs. The skill also enables archiving links and retrieving performance metrics such as views and installs.
Grovs Deep Link Creation Skill
Generate tracking deep links with custom metadata, social previews, and campaign associations for mobile applications. Facilitates precise attribution of traffic and installs through configurable JSON payloads and campaign IDs.
Configure iOS and Android Platform Settings
Configure iOS and Android platform settings including bundle IDs, team IDs, package names, and SHA-256 fingerprints. It also manages redirect rules for App Store, Play Store, and fallback URLs.
Manage and Track Marketing Campaigns
Organise deep links into campaigns to track aggregated marketing metrics such as views, opens, installs, and revenue. This skill enables the creation, listing, and archiving of campaigns to monitor performance across grouped links.
Automated changeset entry creation
An agentic skill that automates the creation of changeset entries by analysing git diffs and logs to determine version bump types and generate changelog descriptions.
Outcome-Weighted Trading Memory Architecture
A cognitive memory architecture for trading agents using Outcome-Weighted Memory (OWM) to prioritise past trades based on outcome, context, and recency. It implements a multi-layer pipeline for managing episodic, semantic, procedural, affec…
Behavioral Risk Management for Trading Agents
Provides behavioural and mathematical risk management for trading agents, including affective state monitoring, position sizing, and detection of emotional drift or overtrading.
Binance Trading Memory Bridge
This skill bridges Binance spot trading events with the TradeMemory Protocol to enable persistent trading memory. It allows agents to journal trades, recall similar historical setups, and detect behavioural biases.
Structured Content Summarisation Skill
Provides structured summarisation for diverse content types, including technical specifications, research papers, and code changelogs. It employs specific templates to maintain technical precision and preserve critical quantitative data.
Statistical Data Analysis and Visualisation
Perform end-to-end data profiling, cleaning, and statistical analysis on CSV, Excel, or JSON datasets. This skill enables automated outlier detection, distribution analysis, and the execution of various statistical tests to uncover patterns…
Structured Multi-Dimensional Code Review
Performs structured, multi-dimensional code reviews focusing on correctness, security, performance, and maintainability. It provides actionable feedback, including critical issues, warnings, and suggestions for code improvements.
Research Analysis and Implementation Planning
Extracts actionable implementation details from research sources like PDFs, notebooks, or text files and records structured findings into an experiment log. It identifies method summaries, requirements, and compatibility analysis to facilit…
Machine-readable experiment progress tracking skill
This skill maintains a machine-readable JSON file to track the real-time status, phases, and activities of an ongoing experiment. It allows external dashboards and CLIs to monitor progress, failures, and current operations through a standar…
Research Method Implementation Skill
Automates the generation of Jupyter notebooks for implementing new research methods, ensuring parity with baseline experiments through consistent data loading and metric computation. It manages dependency updates and logs implementation det…
Experiment Management and Setup Skill
Automates the initialization and bookkeeping of research experiments by setting up directory structures, materialising research sources, and managing JSON-based experiment logs.
Automated Experiment Comparison and Reporting
Automates the comparison of baseline and new implementation results by calculating performance metrics and generating structured JSON reports. It handles the systematic updating of experiment logs, summaries, and comparison datasets.
Experiment Benchmarking and Metric Extraction
Defines comparison metrics and extracts baseline values from notebook outputs to update experiment logs for downstream evaluation.
Analyze Current Baseline Implementation
Extracts model architectures, preprocessing steps, and hyperparameters from existing Jupyter notebooks to create a structured experiment log. It also identifies dependencies and generates a requirements file from the notebook's import state…
Extracting Actionable Research Insights
This skill parses research materials such as PDFs, notebooks, or text ideas to extract method summaries, implementation requirements, and compatibility analysis. It records findings as structured JSON entries within an experiment log to fac…
Machine-readable experiment progress tracker
Maintains a machine-readable JSON file to track the real-time status, phases, and activity of autonomous experiments. It ensures consistent state reporting for dashboards and external monitoring tools.
Automated Experiment Comparison and Reporting
Automates the comparison of baseline and new implementation metrics to generate structured performance reports and update experiment logs.