355 skills · sorted by relevance
Create evaluation drafts for existing skills using actual conversations and user corrections. Produce criteria and repeatable test cases, distinguish evidence from specification assumptions, and prepare an anonymized draft for human review.
Check Screenpipe's process, API, screen capture, and audio transcription status. Inspect logs, storage, resource usage, and processing metrics to identify problems, then report the findings with relevant errors and suggested fixes.
Audit citations in a LaTeX paper against online sources. Verify that references exist, bibliographic details are accurate, and each cited work supports the surrounding claim, then produce reports with findings and proposed corrections.
Review Agent-Native Research Artifacts, structured research packages for agents, across six dimensions of research quality. Produce a JSON report with scores, strengths, severity-ranked findings, practical improvements, and an overall acceptance or rejection recommendation.
Support academic work from refining a research question to literature reviews, manuscript drafting, citation checks, and revision. Select workflows for the current stage, including study selection against eligibility criteria, peer review, and experiment planning.
Check project progress, summarize recent work and remaining tasks, and move to the next GSD workflow step. It also supports automatic step chaining and routes plain-language requests to suitable GSD commands.
Review Renovate pull requests that update GitHub Actions. Verify pinned commits against upstream version tags, inspect release notes for breaking changes, and check existing workflow settings before giving a clear recommendation on whether to merge.
Delegate file creation to shell_agent when direct generation tools fail. Specify the required format and content, then locate generated files by extension or modification time, copy them to the working directory, and check their integrity.
Check generated PDFs with PyPDF2 before reporting completion. A Python script verifies that each file can be opened and has the expected page count, helping identify outputs that need regeneration.
Check generated PDFs with Python and PyPDF2 before delivery. Verify expected page counts, detect unreadable or zero-page files, and collect results across multiple documents, with optional file size checks for unusually small outputs.
Check generated PDFs with PyPDF2, a Python library for reading PDF files. Verify that files exist, contain data, open successfully, and match expected page counts, then use failures to guide regeneration and another verification pass.
Assess API business workflows for flaws that automated scanners can miss. The skill examines skipped steps, simultaneous transactions, authentication tokens, GraphQL, and webhooks, helping testers document weaknesses and their financial or operational impact during authorized assessments.
Tests whether an AI agent accepts unverified plugins, tools, models, or updates. It checks risks introduced through external dependencies, including unsafe installation sources and missing version controls, and reports confirmed findings with conversation evidence.
Guides LoopX benchmark experiments through benchmark-toolkit, covering run management, comparisons under matched conditions, and analysis of agent action histories. It helps explain outcomes and efficiency while tracking source versions, score validity, and evidence boundaries.
Review local skills for description quality, organization, writing style, and file structure. Get weighted scores, letter grades, and concrete improvement plans, or compare multiple skills to decide which issues to fix first.
Review local skills for description quality, organization, writing style, and file structure. Get weighted scores, letter grades, and concrete improvement plans, or compare multiple skills to decide which issues to fix first.
Audit a research replication package with ten checks covering R syntax, file references, output freshness, dependencies, data sources, and documentation. Receive a pass/fail report with specific issues and suggested fixes for preparing the package for submission.
Compare original and rewritten academic text to check that citations, statistics, and core claims remain intact. Receive a report covering writing quality, scholarly tone, and whether the changes should be approved, reviewed, or rejected.
Check a paper's numerical claims against saved analysis outputs before submission. Receive a report identifying verified figures, unsupported claims, missing files, unreferenced outputs, and citation keys absent from the bibliography, without changing the manuscript.
Build and check research replication packages so a project's results can be reproduced. Routes checks to a dedicated agent that examines seeds, versions, file paths, workflow integrity, data documentation, execution environments, and matching outputs.
Prepare survey and administrative datasets for analysis with Python, R, or Stata. Generate cleaning code to handle missing values, duplicates, outliers, and derived variables, with documented transformations and integrity checks that support reproducible research.
Evaluate scientific manuscripts and grant proposals for sound methods, statistics, reproducibility, ethics, and clear reporting. Produce a structured review that separates major concerns from minor issues and gives authors specific, constructive suggestions for revision.
Inspect files with Apache Tika to identify their actual format, embedded content, revision traces, and risk indicators. Save extracted evidence and file hashes while distinguishing recorded claims from observed structure, without asserting authorship or validating signatures.
Update pnpm locally and align package.json with GitHub Actions settings. Check the published package and CI installation paths first, then record package integrity and pin the setup action to a specific commit.