843 skills · sorted by relevance
Keep report writing focused on the original request when information retrieval fails. This skill guides agents to use existing knowledge, disclose missing information, create a structured document with write_file, and verify that the file exists.
Run Python code directly on the host when repeated e2b sandbox failures block execution. The skill explains saving scripts, launching them through the shell, and checking outputs when an isolated environment is unnecessary.
Run Python through the shell when a code sandbox fails to start or connect. Uses a heredoc, a way to pass multiline code, with examples for generating charts, Word documents, and processed JSON files.
Run Python through run_shell when execute_code_sandbox fails unexpectedly. The skill explains how to pass multiline code using a heredoc, a shell syntax for inline text, then check standard output and error messages.
Recover Python file-generation tasks when sandbox execution fails by saving a script and running it through the shell, a command interface. Use the workflow for PDFs, document file operations, dependency checks, and output verification.
Continue file reading and data processing when read_file or execute_code_sandbox fails. The skill uses run_shell with short Python commands or multiline scripts to inspect files, parse data, and save transformed output.
Keep research tasks moving when web search or page reading fails. The workflow identifies tool errors, shifts to existing knowledge where appropriate, and helps produce documents that clearly mark uncertainties and information needing external verification.
Continue regulatory research when official sources cannot be accessed. The skill uses alternative sources or established regulatory knowledge to draft summaries and checklists, while documenting limitations and identifying the need for verification against current official requirements.
Recover failed Python tasks by saving the code as a script and running it through the shell. Provides steps to inspect errors and confirm generated files for spreadsheet creation, data processing, and calculations.
Continue library-dependent Python tasks when sandbox execution fails by switching to a shell script. Covers recognizing import errors, preserving script text, checking execution output, and troubleshooting package availability for tasks such as PDF generation or visualization.
Continue regulatory research when government pages or PDFs cannot be retrieved. Use alternative sources to draft compliance checklists, clearly distinguish the basis of each finding, and identify requirements that need confirmation from the relevant authority.
Extract text from downloaded PDFs by checking for invalid files, trying three extraction tools in sequence, and validating the results. This workflow helps identify download errors, empty output, and encoding problems during document retrieval.
Extract text from difficult PDFs by trying several libraries and command-line tools in sequence, then using OCR, which reads text from images. Record failed attempts and check that extracted content meets the task requirements.
Extract text from PDFs when the first reading attempt fails or produces poor results. The skill tries command-line tools, Python libraries, then OCR, which recognizes text in images, while checking output quality and recording successful methods.
Package project files into a ZIP archive and verify that the deliverable exists and is not empty before completing a task. It also defines a fallback for archiving available files when the agent approaches its iteration limit.
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.
Download PDFs from web URLs or process local files, verify their format, and extract text with pdftotext. If extraction fails or looks poor, try PyMuPDF, check the results, and document any remaining source limitations.
Extract readable text from PDFs when normal file reading returns garbled or unusable content. The workflow switches between extraction tools, checks the results, and includes optical character recognition for scanned pages when needed.
Extract text from local or downloaded PDFs using a sequence of tools. Try file reading, then pdftotext before a Python fallback, check readability, and document failures while clearly separating extracted content from knowledge-based additions.
Download PDFs, check that the files contain valid content, and try several text extraction tools in sequence. The workflow helps detect downloaded error pages, insufficient text, and extraction failures while recording diagnostic details.
Recover from failed Excel creation or editing by switching from execute_code_sandbox to shell_agent. It guides clear task descriptions so the agent can install missing libraries, run code, retry errors, and confirm file creation.
Extract text, tables, and headings from Word documents when standard file reading tools fail. The workflow uses Python and python-docx through run_shell to recover document content for further review or analysis.
Create and check files through shell scripts when code execution tools repeatedly fail. The workflow verifies working directories, handles script quoting and permissions, and inspects outputs such as PDFs and DOCX documents for file size, type, and structure.
Recover from failed file and directory operations by switching to run_shell, a shell command tool. Create nested folders, write files, and verify each result before continuing with tasks that depend on those files.