Menü
Generate and apply structured code changes to a file
Mit Codex oder Claude installieren Kopieren Sie diesen Prompt, fügen Sie ihn in Codex, Claude oder einen anderen Assistant ein und lassen Sie die Skill-Seite prüfen und installieren.
Basierend auf der SOC-Berufsklassifikation
| name | diff |
| description | Generate and apply structured code changes to a file |
| tags | ["editing","code"] |
| tools_required | ["read_file","write_file"] |
| parameters | [{"name":"file","type":"string","description":"Path to the file to modify","required":true},{"name":"changes","type":"string","description":"Description of desired changes","required":true}] |
| output_format | text |
Read a file, plan precise changes based on a natural-language description, and apply them.
Read the file: Use read_file to load the full contents of {file}. Note the total line count and language/format.
Understand context: Before making changes, understand the file's purpose, its imports/dependencies, and how the section you are modifying fits into the broader structure.
Plan changes: Based on the instruction ({changes}), identify the minimal set of edits needed. For each edit, note:
Generate a unified diff: Produce the changes as a unified diff (--- a/file, +++ b/file, @@ line ranges @@). Include enough context lines (3+) for each hunk to make the diff unambiguous.
Apply the changes: Use write_file to write the modified file contents. The written file must be the complete, final version -- not a partial patch.
Verify: Re-read the file with read_file to confirm the write succeeded and the content matches expectations. Check that:
Report: Respond with:
Fix agents that read stale production data instead of classifying direct benchmark prompts — add format detection at top of instructions
Diagnose and fix sub-agent configuration issues in YAML + instruction files that cause timeouts, tool failures, or degraded performance
Fallback persistence paths when primary tools (store_memory, create_note, write_file) are unavailable — memory blocks, task resolution fields, CRM notes, and reporting, in priority order.
Compose and send an email — using native GWS tools (never gog CLI for replies)
Create, advance, and resolve CRM tasks correctly — respecting status transitions, handling retries, and avoiding common API pitfalls.
Audit and calibrate benchmark suite patterns against actual agent output to eliminate false-positives and false-negatives
