Content
82%Weight 40%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
A well-structured, highly actionable body: copy-paste script commands, four concrete workflows, a decision-oriented trigger table, and verified one-level references. Weaknesses are mild — an expendable Keywords section, some body/reference duplication, and missing validation checkpoints in the workflows.
Suggestions
Remove the Keywords section from the body — the frontmatter description already carries the trigger terms, so this section spends tokens without adding guidance.
Trim the inline regulatory trigger table and standard-contracts list to brief summaries that point to references/ip_and_regulatory.md and references/contracts_playbook.md, where the fuller versions already live, to reduce duplication.
Add a validation checkpoint to the workflows (e.g., in Workflow 2, step for handling schema-mismatch errors from term_sheet_analyzer.py, and in Workflow 1, verifying the scanner parsed the contract correctly) to close the feedback-loop gap.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense and assumes domain competence (no padding explaining what an NDA or liquidation preference is in tutorial terms), but the standalone Keywords section and the restated opening line are trimmable, fitting 'efficient; minor instances of over-explanation' rather than score-5's 'every token earns its place'. | 4 / 5 |
Actionability | Quick Start gives copy-paste-ready commands for both bundled scripts with and without arguments, workflows specify exact invocations ("python scripts/term_sheet_analyzer.py path/to/term_sheet.json"), and the Output Standard is a concrete fill-in template — fully executable covering the common cases, matching the score-5 anchor. | 5 / 5 |
Workflow Clarity | Four workflows are clearly sequenced with numbered, specific steps and decision guidance ("Negotiate the worst 3 clauses"), but no explicit validation/feedback checkpoints exist (e.g., verifying scanner output or handling a term sheet that fails schema) — above score-3's implicit checkpoints but short of score-5's explicit validation loops. | 4 / 5 |
Progressive Disclosure | References are one level deep, real (all three files and both scripts exist in the bundle), and well signaled both inline ("See references/ip_and_regulatory.md") and in a footer list; however, the inline regulatory trigger table and standard-contract list substantially duplicate richer tables in ip_and_regulatory.md and contracts_playbook.md — good structure with minor placement gaps, not the clean split of score-5. | 4 / 5 |
Total | 17 / 20 Passed |