Content
57%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.
The body is actionable and reasonably well-sequenced with concrete API code, but it inlines everything into one file with no progressive disclosure to bundle files, embeds time-sensitive pricing without a deprecation context, and lacks explicit validation checkpoints before destructive batch operations.
Suggestions
Add a validate-before-act checkpoint in Step 4 (e.g., confirm account is truly inactive and export a backup before removing) and an audit→review→apply loop so destructive batch operations cannot score above the workflow cap.
Move the full TypeScript audit/optimization scripts and the pricing/API reference into references/ or scripts/ bundle files, keeping SKILL.md as an overview with one-level-deep links to improve progressive disclosure.
Relocate the dated pricing table into a references/ file with a 'verify current pricing at instantly.ai/pricing' note, or place stale figures under an 'old patterns' heading, to remove the time-sensitivity conciseness penalty.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Mostly efficient code with useful comments, but the 'Instantly Pricing Tiers (2026)' table embeds time-sensitive dollar amounts without a deprecation/old-patterns context, which the guidelines penalize; matches anchor 3 rather than 4. | 3 / 5 |
Actionability | Full executable TypeScript with concrete API endpoints covering audit, campaign efficiency, plan sizing, and optimizations, matching anchor 4; not a 5 because the examples depend on an unbundled './src/instantly' helper that is not provided in the skill bundle. | 4 / 5 |
Workflow Clarity | Four steps are clearly sequenced (Audit → Campaign Efficiency → Plan Right-Sizing → Cost Reduction) with a closing checklist, but Step 4 covers batch/destructive operations (removing accounts, archiving campaigns, blocklist edits) without explicit validate-before-act checkpoints, triggering the cap at 3. | 3 / 5 |
Progressive Disclosure | Good section structure (## headers, checklist, error table) but all content is inlined in a single SKILL.md with no references/, scripts/, or assets/ bundle files; the only reference ('instantly-reference-architecture' in Next Steps) points to another skill, not a file, matching anchor 3. | 3 / 5 |
Total | 13 / 20 Passed |