Write large files in segments with anchor lines and batch verification.
npx skills add https://github.com/RuifengFu/agent-skills --skill writing-long-files
Please help me install this Agent Skill. Skill: writing-long-files Source: https://github.com/RuifengFu/agent-skills/tree/main/hermes/writing-long-files Command: npx skills add https://github.com/RuifengFu/agent-skills --skill writing-long-files
Handles the challenge of writing very long files that might fail or get truncated when written as a single piece.
Start by writing the first chunk of the file. Use write_file(path="long-report.md", content="First section of the document...END_OF_SECTION_1"). After writing all chunks, use patch(path="long-report.md", old_string="END_OF_SECTION_1", new_string="END_OF_SECTION_1\n\n...second section...") for subsequent sections.
High-intent search queries and answers about installing and using this skill.
To write large files without truncation or timeouts, write the content in segments using unique anchor lines. Append subsequent batches by patching the file at the anchor, verifying data integrity after each write.
The anchor line technique uses unique text markers to identify where to append new content in a file. You write an anchor like END_OF_SECTION_1, then use a patch function to replace it with the next segment and a new anchor.
Compile large research documents by writing the first chunk to a file, then using a patch method to append subsequent sections at designated anchor lines. This segmented approach bypasses single execution limits for long-form reports.
Yes, segmented file writing is designed for distributed computing environments. It mitigates risks of timeouts and truncation during large-scale data generation by writing and verifying content in manageable batches.
Writing large files as a single piece risks truncation and execution timeouts. When file operations exceed single execution limits, the process may fail or generate incomplete text files, requiring segmented writing to ensure data integrity.
wxyCreate files via create_file and append segments for large outputs.
jenswittmannAutomate project builds and asset compilation in CI/CD pipelines.
damilola-elegbede-orgCreate, edit, and extract slide content from PPTX presentations.
cas-bigdatalabEliminate duplicate images in large datasets using Ray distributed computing.
