226 skills · sorted by relevance
Conducts privacy auditing of AI models including training data extraction testing, membership inference attacks, model inversion testing, and attribute inference assessment. Uses ML Privacy Meter and related tools to quantify privacy leakage. Keywords: model audit, membership inference, privacy meter, model inversion, training data extraction.
The English summary is being prepared. Showing the original description.
Instruments code to track the flow of untrusted or sensitive data at runtime, enabling detection of injection vulnerabilities, data leaks, and privilege violations. Use when users need to: (1) Track untrusted input propagation through code, (2) Detect SQL injection, XSS, or command injection vulnerabilities, (3) Identify sensitive data leaks, (4) Monitor privilege escalation paths, (5) Perform dynamic taint analysis for security testing. Supports Python, Java, JavaScript, and C/C++ with configurable taint sources and sinks.
The English summary is being prepared. Showing the original description.
Supabase security expert to audit RLS (Row Level Security), RBAC, relational databases, prevent data leakage, and utilize Supabase Linter / Ahli keamanan Supabase untuk audit RLS (Row Level Security), RBAC, database relasional, pencegahan kebocoran data, dan pemanfaatan Supabase Linter.
The English summary is being prepared. Showing the original description.
Firebase security expert to audit Security Rules (Firestore/Realtime Database/Storage), authentication, API keys, data leakage prevention, and App Check configuration / Ahli keamanan Firebase untuk audit Security Rules (Firestore/Realtime Database/Storage), autentikasi, API keys, pencegahan kebocoran data, dan konfigurasi App Check.
The English summary is being prepared. Showing the original description.
Audit existing model training code — find reproducibility issues, data leakage, and missing best practices. Use when asked to "audit our training code", "is our training reproducible", or "check for training data leakage".
The English summary is being prepared. Showing the original description.
Use Therapeutics Data Commons through the PyTDC Python package for registry discovery, approved dataset access, task-aware splits (scaffold, cold-start, temporal, combination), evaluator metrics, benchmark groups, and bounded molecular-oracle workflows. Use this skill to find which TDC datasets exist for a therapeutic task, load them with a split that does not leak, score predictions with the task's own official metric rather than a generic one, and run benchmark groups reproducibly. Also trigger on PyTDC, Therapeutics Data Commons, tdc.single_pred, tdc.multi_pred, ADMET Benchmark Group, scaffold split, get_split, or molecular oracles such as GSK3B, JNK3 and DRD2.
The English summary is being prepared. Showing the original description.
Decide what to PERSIST in your data structures (graph nodes, DB tables, cache layouts) based on the QUERIES you must answer, not the DEFINITIONS that justify the data. Definitions tell you logical identity; storage strategy is independent. Materializing every definitional component leads to O(N×M) blow-ups that masquerade as memory leaks. Use when: (1) designing a new graph-node type, edge type, DB table, or cache layout; (2) about to name a `subgraph` / `comprises` / `members` / `contains` collection on a parent entity; (3) memory growth proportional to (entities × per-entity-elements); (4) "memory leak" debugging that survives multiple targeted patches — the leak may be storage shape, not retention; (5) pre-implementation sketch includes "for each X, store all reached nodes"; (6) research/spec doc says "X = ⟨A, B, C⟩" and you're tempted to make it three storage edges per X.
The English summary is being prepared. Showing the original description.
Monitors dark web forums, marketplaces, paste sites, and ransomware leak sites for mentions of organizational assets, leaked credentials, threatened attacks, and threat actor communications to provide early warning intelligence. Use when establishing dark web monitoring coverage, investigating specific data breach claims, or enriching incident investigations with dark web context. Activates for requests involving dark web OSINT, leak site monitoring, credential exposure, Recorded Future dark web, or Tor hidden service intelligence.
The English summary is being prepared. Showing the original description.
- Check and interpret data-breach exposure for an email, username, phone or name using Have I Been Pwned, the Pwned Passwords k-anonymity range API, DeHashed, IntelX and Snusbase. Use when checking breach or leak exposure, finding which services an account was registered with, interpreting a combolist or credential dump, assessing credential compromise, or auditing your own leaked personal data
The English summary is being prepared. Showing the original description.
Plan and review machine learning systems for production use, covering data requirements, repeatable training, quality checks, deployment, monitoring, and rollback. Produce concrete plans and review findings for moving model experiments into operational systems.
Create, edit, and clean spreadsheet files, including Excel workbooks and CSV data. Build formulas and formatting with openpyxl, a Python library, then recalculate and validate workbooks to check for formula errors and file problems.
Search contacts and calendar events already stored in Muse when a paired device is unavailable. Retrieve relevant records, check whether results are complete, or delete Muse's local contact or calendar copies without changing data on the devices.
Build and check clinical machine learning workflows with PyHealth, from loading electronic health records, signals, or images to training and evaluation. Includes patient-level data splits, prediction task checks, model checkpoints, and medical code mapping.
Measure Python execution time and memory use to locate slow or resource-heavy code. Use profiling results to guide improvements in data structures, caching, input and output operations, algorithms, and parallel processing.
Analyze proxy, DNS, and firewall traffic logs to uncover unapproved SaaS and cloud services. Group usage by domain, compare it with approved applications, and produce a report with traffic volumes, user counts, risk scores, and remediation recommendations.
Investigate suspicious mobile apps by examining permissions, network traffic, and runtime behavior. Use observed evidence to identify signs of data theft, SMS interception, credential stealing, or communication with attacker control servers, then classify the suspected malware.
Map an organization's internet-facing assets using public information, including domains, indexed services, employee details, and leaked data. Compile findings into a reconnaissance report to support authorized security assessments and identify potential exposure.
Write and review .NET calls to native C/C++ libraries using P/Invoke. Covers function declarations, string conversion, memory ownership, and platform differences to help investigate crashes, memory leaks, and corrupted data.
Test LLM applications for malicious instructions, information leaks, and unsafe tool use. The skill runs attack cases and checks observable evidence, helping distinguish suspicious replies from reproducible failures with concrete impact.
Investigate LDAP and XPath injection in directory searches and XML-backed authentication. The skill guides checks for login bypasses and information leaks, uses control responses to reduce false positives, and distinguishes Active Directory from other LDAP directories.
Assess GraphQL APIs in financial services for unauthorized transfers, rounding errors, duplicate transactions, and exposed personal data. The skill guides checks of account permissions and ledger changes to distinguish actual financial impact from misleading success responses.
Assess Kubernetes and Docker security across exposed APIs, access permissions, service account tokens, and container boundaries. Use actual data reads or command output to establish impact and distinguish information exposure from broader compromise.
Optimize Python code for reduced memory usage and improved memory efficiency. Use when asked to reduce memory footprint, fix memory leaks, optimize data structures for memory, handle large datasets efficiently, or diagnose memory issues. Covers object sizing, generator patterns, efficient data structures, and memory profiling strategies.
The English summary is being prepared. Showing the original description.
Think and work like an expert AI Researcher. Use when a task calls for AI Researcher judgment. Reasons from data generating processes, inductive biases, and compute-data-algorithm trade-offs through train/val/test discipline, seed sweeps, ablation ladders, and standards like NeurIPS reproducibility checklists, model cards, and lm-eval-harness, while treating data leakage (Kapoor & Narayanan taxonomy), benchmark contamination, spurious correlations, and reward-model overoptimization as first-class failure modes.
The English summary is being prepared. Showing the original description.