Content
86%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 dense, well-structured body that assumes competence, leads with executable commands and concrete formulas, and uses a clear 10-step workflow backed by validation scripts and a release-gate checklist. Main gaps are the absence of explicit per-step fix-and-retry feedback loops and some long bulleted lists that could be tightened or offloaded to references.
Suggestions
Add explicit feedback-loop language at each validation step (e.g., 'If validate_evidence_ledger.py reports errors, fix the ledger and re-run until clean') to lift workflow_clarity from 4 to 5.
Move the longer enumerations (research-contract clarifications, source-ledger field list, measurement guardrails) into the existing references and keep only the operative subset inline, to improve conciseness.
Cross-link each workflow step to its named reference (e.g., step 5 → data_analysis_patterns.md, step 8 → methods_and_ethics.md) in-line rather than only at the end, to make navigation even more direct.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Efficient and assumes Claude's competence — it never explains concepts Claude already knows (no 'what is TAM' or 'what is HHI') — but the ~325-line body carries several extensive bulleted lists (research contract, source-ledger fields, measurement guardrails) that could be trimmed or moved to references, keeping it just below the 'every token earns its place' 5 anchor. | 4 / 5 |
Actionability | Provides copy-paste-ready, executable commands covering the common cases (validate_evidence_ledger.py, audit_claim_citations.py, calculate_market_sizing.py, forecast_sensitivity.py, validate_competitor_matrix.py, check_unit_consistency.py, generate_report_scaffold.py) plus concrete formulas (TAM_top, TAM_bottom, SAM_s, SOM_s) and specific ID conventions (S-001, C-001), matching the 'fully executable, copy-paste ready' anchor. | 5 / 5 |
Workflow Clarity | A clearly sequenced 10-step workflow with explicit validation checkpoints (the validate scripts) and a final 'Release gate' checklist; it stays at 4 rather than 5 because the per-step 'validate → fix → re-validate' feedback loops are implied by the gate rather than spelled out as in the 5 anchor, and validation is present so the destructive/batch cap of 3 does not apply. | 4 / 5 |
Progressive Disclosure | The body is a clear overview pointing to well-signaled, one-level-deep references (references/*.md), templates (assets/*), and tools (scripts/*) — all verified to exist as real files — with a 'Bundled resources' section giving one-line descriptions for easy navigation and no nested reference chains, matching the 5 anchor. | 5 / 5 |
Total | 18 / 20 Passed |