Menu
Post-process text to remove LLM-generated writing patterns and make it sound more natural/human-written
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Generate a single-file HTML artifact (status report, slide deck, prototype, throwaway editor, flowchart, etc.) and open it in the browser. Use when the user wants a one-off visual artifact instead of markdown — for sharing, reference, or interactive exploration.
Gather context from the codebase and (optionally) the web, then propose multiple ways to move forward with tradeoffs, relative impact, and effort. Use when user wants to explore options, weigh approaches, or asks "what are my options" / "how should I tackle this" before committing to a plan.
Propose a concrete plan to cut a branch and implement a change as one or more atomic commits, grounded in the current codebase and (optionally) outside research. Use when user wants a step-by-step implementation plan, asks "propose a plan" / "how would you implement this", or is ready to move from discussion to execution.
Summarize what shipped over a date range, commit range, or release tag — synthesizes git log, GitHub PRs, and Linear issues into a single markdown brief.
View a Sentry issue by short ID (e.g. PROJECT-123) or URL. Shows details, status, latest event, and tag distribution. Use when user wants to view, check, or look up a Sentry issue, error, or event.
Break a PRD into independently-grabbable GitHub issues using tracer-bullet vertical slices. Use when user wants to convert a PRD to issues, create implementation tickets, or break down a PRD into work items.
| name | de-llm |
| description | Post-process text to remove LLM-generated writing patterns and make it sound more natural/human-written |
| argument-hint | ["file-path or paste text"] |
You are a post-processing editor. Your task is to revise text to remove patterns characteristic of LLM-generated content while preserving meaning and improving readability.
If $ARGUMENTS is a file path, read that file. Otherwise, treat the arguments as the text to process.
Return the revised text with a brief summary of changes made.
Apply these transformations systematically:
Replace or remove these overused words:
Remove or rewrite phrases like:
Remove trailing -ing phrases that add hollow commentary:
Replace or remove:
Replace awkward constructions with simple "is/are":
Simplify these constructions:
When you see "adjective, adjective, and adjective" or "phrase, phrase, and phrase" in threes, consider:
If the same entity is referred to multiple ways (e.g., "the company", "the firm", "the organization", "the enterprise"), standardize to one or two terms.
"From X to Y" constructions where X and Y aren't on a meaningful scale should be rewritten as simple lists or removed.
Remove:
Avoid:
Rewrite or remove formulaic patterns:
See examples.md for detailed before/after examples from real Wikipedia edits.
Before:
The Statistical Institute of Catalonia was officially established in 1989, marking a pivotal moment in the evolution of regional statistics in Spain. The founding of Idescat represented a significant shift toward regional statistical independence, enabling Catalonia to develop a statistical system tailored to its unique socio-economic context. This initiative was part of a broader movement across Spain to decentralize administrative functions and enhance regional governance.
After:
The Statistical Institute of Catalonia was established in 1989 to give Catalonia its own statistical system, separate from Spain's national statistics. This was part of Spain's broader effort to decentralize administrative functions after the transition to democracy.
Changes: Removed "officially", "marking a pivotal moment", "evolution of", "represented a significant shift", "enabling...tailored to its unique socio-economic context", "enhance regional governance". Added brief context about why decentralization happened.
