Content
62%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The body is richly structured with a clear workflow and validation, but it is verbose with redundant description repetition, and its code examples are misleading — they reference a module/class/script that do not match the actual bundled `main.py`.
Suggestions
Reconcile code examples with the real bundle: either implement `GraphInterpreter`/`graph_interpreter.py` with the shown methods, or rewrite examples to use the actual `GraphInterpretation` class and `describe()` method in `scripts/main.py`.
Remove the verbatim repetition of the description in 'When to Use' and 'Key Features'; reference it once or summarize.
Make `references/` navigation explicit by linking `guidelines.md` by name, and either populate it with the real statistical-reporting/figure-caption guidance or drop the reference.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Mostly actionable but padded: the frontmatter description is repeated verbatim in both 'When to Use' and 'Key Features', and Quick Start / Core Capabilities / Common Patterns overlap with similar interpreter code; could be tightened considerably. | 2 / 3 |
Actionability | Code looks concrete but is not executable against the actual bundle: examples import `from scripts.graph_interpreter import GraphInterpreter` and call `python scripts/graph_interpreter.py`, yet the only script is `scripts/main.py` defining class `GraphInterpretation` with a single `describe` method — the referenced module, class, and methods do not exist. | 2 / 3 |
Workflow Clarity | A 5-step Workflow is sequenced with explicit validation ('stop early if the task would require unsupported assumptions'), a Quick Check (`python -m py_compile`), a fallback path on failure, and a Quality Checklist — matching the anchor for clear sequence with checkpoints and feedback loops. | 3 / 3 |
Progressive Disclosure | Structure exists but references are poorly signaled: the body points generically to `references/` (never naming `guidelines.md`, which itself is near-empty) and references a non-existent `scripts/graph_interpreter.py`, while ~400 lines of API-style material sit inline that could live in separate files. | 2 / 3 |
Total | 9 / 12 Passed |