383 skills · sorted by relevance
Organize repeated experiments to improve models, prompts, or agent systems against measurable goals. The skill records hypotheses and lessons in a persistent tree, isolates trials, and compares promising candidates using evaluation data kept separate from development.
Design layered protections for AI agents against malicious instructions, tool misuse, and data leakage. Provides guidance and code examples for input checks, isolated tool execution, permission limits, and audit logging to help review and strengthen agent security.
Draft Prisma 8 release notes from merged PRs since the previous published release. Select user-facing changes, put breaking changes first, link migration guidance and PR evidence, and mirror the notes in CHANGELOG.md.
Prepare the next Prisma 8 release by updating workspace versions, assembling upgrade guides, and drafting release notes. Open release and documentation pull requests for maintainer review; merging the release request triggers automated publication.
Set up distributed tracing with Jaeger and Tempo to follow requests across services. Examine delays, service dependencies, and error propagation, with guidance on sampling, linking logs to traces, and troubleshooting missing trace data.
Design application and API behavior when operations fail. Choose between exceptions and Result types, which explicitly represent success or failure, and improve retries, error messages, logging, and resource cleanup so failures are easier to handle and investigate.
Build an Excel comparison of public companies using financial performance, valuation ratios, and statistical benchmarks. The spreadsheet includes formulas, source notes, medians, and quartiles to show how each company compares with its peers.
Use Curviate CLI to retrieve LinkedIn profiles and company information, update your own profile, inspect company inboxes, and convert search terms into filter IDs. It also covers account connections and choosing stored or live profile data.
Manage LinkedIn connection requests and follows with the Curviate CLI. Review pending invitations, preview actions, withdraw requests, and list connections or followers, with guidance on duplicate requests and delays before invitations appear.
Answer questions from saved gbrain knowledge using keyword, meaning-based, and relationship searches. The skill combines relevant evidence with page citations and highlights missing information, conflicting sources, or potentially outdated results.
Turn videos, audio, PDFs, books, screenshots, and GitHub repositories into knowledge pages with summaries and highlights. Preserve original files, extract people and companies, and connect the material to their pages and timelines for later reference.
Turn meeting transcripts into structured GBrain pages with decisions, assigned tasks, and verified quotes. Resolve speakers using evidence, split recordings containing separate meetings, and update related people and companies with links and dated meeting context.
Prepare for a GDPR audit with six structured questions covering processing records, lawful grounds, privacy risks, individual rights requests, international transfers, and breaches. Organize findings with article references and identify the next three actions.
Develop and modify the Handsontable wrapper for Vue 3 with guidance on component structure, settings updates, column configuration, and shared data references. It also explains the build order and tests used to check wrapper changes.
Manage credentials for Git access and deployments in KubeSphere DevOps. Covers SSH keys, passwords, API tokens, and Kubernetes connection settings, including their use in Jenkins pipelines and troubleshooting synchronization failures.
Improve GSAP animation performance by reviewing animated properties, browser layout work, and simultaneous updates. Use transforms, grouped updates, reusable animations, and targeted cleanup to reduce stuttering in lists, mouse followers, and scroll effects.
Recover PDF or DOCX generation when execute_code_sandbox fails with unclear errors. Use run_shell to execute Python directly, check required libraries, report specific failures, and verify that the output file was created.
Recover document and PDF generation tasks when sandbox errors provide little detail. Checks Python libraries, runs generation through the shell, captures errors, and verifies output files, with alternative PDF tools and text output as fallback options.
Compare code structure across two commits, tags, or directories to identify security-relevant changes. Produces a report on new attack paths, privilege boundaries, complexity, and how widely a change could affect other functions.
Investigate how a limited Linux account could gain root, the system administrator's privileges. The skill guides checks of permissions, scheduled tasks, services, and known vulnerabilities, while outlining evidence defenders can monitor during authorized security assessments.
Check whether a single tool outage, invalid input, or repeated retry can disrupt an entire agent workflow. Use dialogue tests to identify spreading errors, missing fallbacks, and uncontrolled retries, then document confirmed findings.
Give navigation, controls, and feature cards a consistent icon style using Iconify Solar Duotone Bold. It guides icon sizing, clear labels, two-tone opacity, and contrast so icons fit the surrounding interface.
Apply subtle gradient borders to cards, pricing panels, buttons, and other web surfaces. Choose layered CSS backgrounds or a masked border to suit existing designs, with guidance on Tailwind syntax, corner alignment, opacity, and light and dark themes.
Build a scanning entrance for Three.js models, revealing their wire structure before the solid surface appears. Coordinate both layers with one expanding wave, then remove the temporary wires when the animation finishes.