Menu
Run a prompt or instruction across multiple files matching a glob pattern
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 and apply structured code changes to a file
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.
| name | batch |
| description | Run a prompt or instruction across multiple files matching a glob pattern |
| tags | ["automation","bulk"] |
| tools_required | ["spawn_agents","list_directory","read_file"] |
| parameters | [{"name":"glob","type":"file_glob","description":"File pattern to process (e.g., \"src/**/*.py\")","required":true},{"name":"instruction","type":"string","description":"What to do with each file","required":true},{"name":"concurrency","type":"integer","description":"Max concurrent sub-agents","default":3}] |
| output_format | json |
| composable | false |
Run an instruction against every file matching a glob pattern, using sub-agents for parallel execution.
Expand the glob: Use list_directory to walk the directory tree implied by {glob}. Collect the full list of matching file paths. If zero files match, return immediately with {"files_processed": 0, "error": "no files matched pattern"}.
Preview: Log the total count of matched files. If more than 50 files match, warn the user and ask for confirmation before proceeding (unless running autonomously).
Read and validate: For each file, use read_file to confirm it exists and is readable. Skip binary or unreadable files and note them in a skipped list.
Spawn sub-agents: Use spawn_agents to fan out processing. Pass each sub-agent:
{instruction}{concurrency} simultaneous agents.Collect results: As each sub-agent completes, capture its output. Track successes, failures, and any errors with the originating file path.
Summarize: Return a JSON object:
{
"files_processed": <int>,
"succeeded": [{"file": "...", "result": "..."}],
"failed": [{"file": "...", "error": "..."}],
"skipped": ["..."]
}
warning field in the output suggesting the instruction may need adjustment.