Content
87%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
An actionable, token-efficient reference with concrete commands and well-organized tables. Its main weakness is workflow clarity: the verification step is awkwardly ordered, lacks an error-recovery loop, and is absent for import.
Suggestions
Reorder so 'Verify Export' follows the Export section, and frame it as a numbered sequence: export → verify → if output doesn't match, fix flags and re-verify.
Add a parallel verify step for Import (e.g., list the created agent.yaml/SOUL.md files and inspect one) so both directions have a checkpoint.
Add an explicit error-recovery note for the verify step (what to check and re-run when the output is missing or wrong) to turn the checkpoint into a true feedback loop.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Lean throughout — compact format/source tables and commented command examples with no padding or explanation of concepts Claude already knows; every section earns its place. | 3 / 3 |
Actionability | Provides fully executable, copy-paste-ready commands such as 'opengap export -f claude-code -d ./my-agent -o CLAUDE.md' with concrete format flags and real example invocations for both export and import. | 3 / 3 |
Workflow Clarity | A 'Verify Export' checkpoint exists, but it is placed before the Export section, has no error-recovery feedback loop, and Import has no corresponding verification step — checkpoints are present but loosely sequenced. | 2 / 3 |
Progressive Disclosure | A single self-contained file organized into clear sections (Verify, Export, Formats, Examples, Import, Sources) with no nested references and no content that needs splitting out. | 3 / 3 |
Total | 11 / 12 Passed |