Content
92%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.
Excellent instruction-only content: lean, concrete, quantified, with clear two-sided workflows and an explicit pre-LOI verification loop that delegates to sibling skills instead of duplicating them. The single significant defect is progressive disclosure: two referenced bundle files (references/integration-playbook.md, references/due-diligence-checklist.md) do not exist, so the skill promises material it cannot deliver.
Suggestions
Create the referenced bundle files references/integration-playbook.md and references/due-diligence-checklist.md, or remove the pointers and inline whatever content is essential — currently the Integration Frameworks and Resources sections promise material that does not exist in the bundle.
Verify the sibling-skill paths (../general-counsel-advisor/scripts/term_sheet_analyzer.py, ../chief-data-officer-advisor/scripts/ai_training_data_audit.py, etc.) resolve in the deployed skill fleet, since the Verification Loop depends on them; note in the body what to do if the sibling tooling is unavailable.
The Keywords section duplicates trigger terms already present in the description frontmatter — fold the few unique terms (LOI, term sheet, earnout, acqui-hire) into the description and drop the section to save tokens.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean and dense — checklists, tables, and quantified red flags ('>30% revenue from 1 customer', '$1-3M per engineer') with zero explanation of concepts Claude already knows. Every section adds non-obvious value (e.g., naming the specific sources to verify multiples against), matching the 'every token earns its place' anchor. | 5 / 5 |
Actionability | Highly actionable for an instruction-only skill: quantified red-flag thresholds, named verification sources with a 'cross-check at least two' rule, a numbered preparation sequence, and exact sibling-script paths to run ('term_sheet_analyzer.py', 'ai_training_data_audit.py') with pass/fail conditions ('Any 🔴 finding goes to outside counsel before signing'). | 5 / 5 |
Workflow Clarity | Both Quick Start sequences are explicit (rationale → screening → DD → valuation → negotiation → integration; readiness → data room → advisor → negotiation → transition), and the 'Verification Loop (before any LOI or signature)' adds explicit validation gates and a feedback loop ('Loop the findings back into the negotiation-points table before the next counter') plus checklists for the DD domains — matching the top anchor. | 5 / 5 |
Progressive Disclosure | Section structure and reference signaling are good, but the body cites 'references/integration-playbook.md' and 'references/due-diligence-checklist.md' (both in the Integration Frameworks section and again in Resources) and neither file exists in the bundle — the references/ directory is absent entirely. Scoring against the actual bundle structure, navigation breaks at the point of use, which is a bigger defect than the 'minor organization gaps' of anchor 4 but the structure is still far better than anchors 1-2. | 3 / 5 |
Total | 18 / 20 Passed |