Content
38%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 Core Capabilities section is genuinely useful (headline formulas, about-section structure, keyword table, experience formula), but the body is padded with duplicated boilerplate, its code and CLI examples do not match the actual `scripts/main.py`, and all three listed reference files are missing from the bundle. Alignment between the documented and actual package is the dominant weakness.
Suggestions
Sync the documented API to the real script: either implement generate_headline, write_about_section, suggest_keywords, and optimize_experiences plus the CLI flags in scripts/main.py, or rewrite the Quick Start/Core Capabilities examples to use the existing optimize(role, specialty, achievements) signature and the correct module path scripts.main.
Fix the reference navigation: create references/linkedin-examples.md, references/keywords-by-specialty.json, and references/headline-templates.md (or repoint those sections to the existing references/guidelines.md and reference it from the body).
Cut the boilerplate: remove the duplicated 'When to Use' description echo, merge Quick Check/Audit-Ready Commands into one section, and drop generic scaffold sections (Implementation Details self-reference, Response Template, Output Requirements) that add no domain-specific value.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | "Use this skill when the task needs Use when optimizing LinkedIn profiles..." verbatim-repeats the description, "Audit-Ready Commands" repeats the same `python -m py_compile scripts/main.py` line already shown in Quick Check and Example Usage, and sections like "Output Requirements", "Response Template", "Implementation Details" ("See `## Workflow` above for related details.") are generic padded boilerplate. Several unnecessary duplicated/padded sections match anchor 2; it is not a 1 because the Core Capabilities sections carry genuine, non-redundant domain content. | 2 / 5 |
Actionability | The Core Capabilities code (`optimizer.generate_headline(...)`, `write_about_section(...)`, `suggest_keywords(...)`, `optimize_experiences(...)`) and CLI (`python scripts/linkedin_optimizer.py --role ...`) look concrete, but the bundled `scripts/main.py` only implements `optimize(role, specialty, achievements)` with no CLI flags, and no `scripts/linkedin_optimizer.py` exists — the examples are not executable as written. Concrete-but-broken guidance with missing key details matches anchor 3; it is not a 4 precisely because none of the shown entry points run against the actual bundle. | 3 / 5 |
Workflow Clarity | There are three overlapping step lists ("Example run plan", "Workflow", "Response Template") with sequence present but the only validation is a syntax check (`py_compile`), and checkpoints are implicit ("Review the generated output"). This matches anchor 3 — steps listed, validation gaps; not a 4 because the duplicated workflows and absent output-validation steps leave real gaps. | 3 / 5 |
Progressive Disclosure | The References section points to `references/linkedin-examples.md`, `references/keywords-by-specialty.json`, and `references/headline-templates.md`, none of which exist, while the actual bundle file `references/guidelines.md` is never referenced. Navigation is effectively broken against the real bundle structure, matching anchor 2 (minimal/broken structure); it is not a 3 because signaling to real files — the core requirement — fails entirely. | 2 / 5 |
Total | 10 / 20 Passed |