orq-improve-agent
Improve an underperforming orq agent, deployment, or local agent — rewrite its instructions against a structured prompting framework, or move a configuration knob, grounded in the error-analysis file orq-analyze-traces writes. Use when a prompt needs improvement, when a config knob is wrong (truncated answers, iteration caps, sampling), or when you have a failure taxonomy and want the fix applied. Do NOT use to re-architect a pipeline (use orq-build-agent), to align a judge that already exists (use orq-evaluator-alignment), or to build the failure taxonomy in the first place (use orq-analyze-traces).
Pinned to revision 495b5ad24c78, so it is the text this page describes rather than whatever the author pushed since.
Pre-approved tools experimental
Experimental field. Support varies between clients, so this list is what the author declared, not what your client will enforce.
- Read
- Write
- Edit
- Grep
- Glob
- Task
- AskUserQuestion
- Bash(orq traces list-fields:*)
- Bash(orq traces list-facets:*)
- Bash(orq traces list-facet-values:*)
- Bash(orq traces aggregate:*)
- Bash(orq traces search:*)
- Bash(orq traces get-span:*)
- Bash(orq traces list-spans:*)
- Bash(orq traces thread:*)
- Bash(orq reporting query:*)
- Bash(orq agents retrieve:*)
- Bash(orq agents get-response:*)
- Bash(orq deployments get-config:*)
- Bash(orq agents update:*)
- Bash(orq tools retrieve:*)
- Bash(orq knowledge-bases retrieve:*)
- Bash(orq memory-stores retrieve:*)
- Bash(orq evals get:*)
- Bash(orq skills get:*)
- mcp__orq-workspace__search_entities
- mcp__orq-workspace__get_agent
- mcp__orq-workspace__get_span
- mcp__orq-workspace__get_deployment
- mcp__orq-workspace__list_traces
- mcp__orq-workspace__search_docs
Files
Every link opens the file at its source, pinned to the revision this page describes.