Content
67%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, largely actionable skill body with clear sections, concrete commands, and useful error handling. Its main weakness is token efficiency: repeated free-tier limits and a redundant Example Workflows section, plus undefined MCP parameters, keep it from the top band.
Suggestions
State the free-tier rate limit (1000 req/day, 60 req/min) once — in Error Handling — and remove the duplicates from the Key Differences table and Authentication section.
Cut or compress the Example Workflows section: it repeats tool syntax already shown in Usage Patterns; keep one or two workflows as concrete one-liner tool invocations instead of narrative blocks.
Define the MCP parameters used in Best Practices (working_dir, timeout_secs, sandbox, yolo_mode) where the tools are first listed, so the guidance is self-contained.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is mostly efficient but includes redundancy that could be tightened: the '1000 req/day free' limit appears three times (Key Differences table, Error Handling, Authentication), and the four 'Example Workflows' blocks largely re-illustrate tool syntax already shown in Usage Patterns using non-executable narrative wrappers like 'I'll use Gemini for real-time search:'. | 3 / 5 |
Actionability | Mostly executable guidance: real CLI commands ('gemini -p "Explain this error message" --output-format json'), a pipe example, MCP tool names, a JSON config snippet, env vars, and Rust code. Minor gaps: the pseudo-syntax '``\n[gemini_search]\nquery: ...\n``' blocks are not copy-paste ready, and parameters like 'working_dir', 'timeout_secs', 'sandbox: true', and 'yolo_mode' are referenced in Best Practices without definition. | 4 / 5 |
Workflow Clarity | The flow is clear and well-sequenced for a delegation skill (When to Use → Usage Patterns → Configuration → Error Handling → Authentication), and the Error Handling section provides explicit failure-recovery guidance ('Not Found: Install with npm install -g @google/gemini-cli', 'Auth Failed: Run gemini interactively'). Minor gaps: no verification step that a Gemini response actually returned before relying on it. | 4 / 5 |
Progressive Disclosure | Good structure with clear section headers and everything at the top level (no nested references, and no references/ or scripts/ bundle files exist to point to). Minor organization gap: at ~160 lines, detailed material like the three Authentication methods and the Example Workflows could be split into reference files to slim the main SKILL.md. | 4 / 5 |
Total | 15 / 20 Passed |