toolbox

Build and run lightweight regression evaluations for agents, tools, prompts, and workflows. Use when comparing agent versions, catching behavior drift, measuring pass rates over repeated runs, or turning production failures into replayable test cases.

Audit agent repositories, MCP configurations, tool schemas, prompts, and logs for common security risks: secret exposure, prompt injection payloads, tool poisoning, command execution, broad filesystem access, and unsafe auto-approval.

Inspect agent traces, tool-call logs, and OpenTelemetry/OpenInference-style span exports. Use when debugging why an agent skipped a tool, called the wrong tool, hit errors, exceeded latency budgets, or produced an unsupported answer.

Design and validate graph/state-machine agent workflows with explicit starts, terminal states, guarded cycles, human approval points, and side-effecting tool boundaries. Use when building LangGraph-style, multi-agent, or durable agent workflows.

Safely list, create, and extract zip/tar archives with path traversal checks. Use when packaging artifacts, unpacking downloaded files, or inspecting archive contents on a user's computer.

Build, run, and review browser-using AI agents with Playwright, Browser Use, MCP browser tools, or hosted browser runtimes. Use for web navigation agents, form-filling, UI task automation, screenshots, DOM/vision checks, auth/session handling, and browser-task reliability reviews.

Read from and write to the system clipboard on macOS or Linux when clipboard commands are available. Use for transferring short text snippets between agent work and the user's desktop applications.

Build source-grounded context packs with file hashes, line ranges, chunk text, and simple retrieval selection. Use when preparing reliable context for agents, RAG experiments, codebase analysis, or long-document tasks where provenance and source traceability matter.

Preview CSV or TSV files in a compact table with delimiter sniffing. Use when inspecting tabular text data before transforming or importing it.

Compute stable checksums for one or more files using hashlib algorithms. Use when the user needs reproducible content verification or artifact IDs.

Plan and optionally apply local file organization tasks: group files by extension, date, or media type, and find duplicate files by hash. Use when a user asks to clean up Downloads, Desktop, project artifacts, or duplicate files.

Validate, summarize, pretty-print, and extract simple dot paths from JSON files or stdin. Use for API responses, config files, and structured data.

Design, lint, and review Model Context Protocol (MCP) tools, resources, prompts, and server configurations. Use when creating MCP servers, exposing APIs to agents, reviewing tool schemas, or checking tool metadata for safety and usability before publishing.

Inspect local processes and network ports to debug dev servers, port conflicts, hung processes, and common localhost issues. Use before killing processes or changing ports; this skill is read-only by default.

Map a repository or directory with a compact file listing, extension summary, and total size. Use before modifying unfamiliar projects or when the user asks for a quick project overview.

Capture desktop screenshots on macOS or Linux using available system commands. Use when the user asks to inspect visible UI state, preserve evidence, or debug desktop/browser rendering outside a controlled Playwright session.

Inspect a user's computer environment: OS, architecture, Python/runtime paths, common developer commands, disk space, selected environment variables, and network identity. Use before installing tools, debugging setup problems, or deciding which local commands are available.

Search project files for text or regex patterns, preferring ripgrep when available and falling back to a Python implementation. Use for fast codebase and document text discovery.

Fetch HTTP or HTTPS URLs, preview response metadata and text, or save the body to disk. Use for lightweight web/API retrieval when JavaScript execution is not required.

Post messages to a social media matrix — X (Twitter), LinkedIn, Facebook Page, Reddit, and Buffer — simultaneously or selectively from a single command. Use when the user wants to publish content to one or more social platforms at once, cross-post announcements, or automate social media publishing workflows.

Command-line utility for common PDF operations: merge all PDFs in a directory into one file (with optional PPTX conversion and title pages), and extract specific pages from a PDF into a new file. Use when the user needs to combine, split, or reorganize PDF documents.