firebase-ai-logic-basics
Official skill for integrating Firebase AI Logic (Gemini API) into web applications. Covers setup, multimodal inference, structured output, and security.
56
33%
Does it follow best practices?
Impact
100%
1.42xAverage score across 3 eval scenarios
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./skills/firebase-ai-logic-basics/SKILL.mdQuality
Discovery
32%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
The description identifies its domain (Firebase AI Logic / Gemini API) and lists topic areas it covers, but it reads more like a table of contents than an actionable skill description. It lacks a 'Use when...' clause, concrete action verbs, and sufficient natural trigger terms that users would employ when requesting help with this technology.
Suggestions
Add an explicit 'Use when...' clause, e.g., 'Use when the user wants to add Gemini AI capabilities to a web app using Firebase, or mentions Firebase AI Logic, Vertex AI in Firebase, or Google AI in a frontend context.'
Replace topic labels with concrete actions, e.g., 'Configures Firebase AI Logic SDK, sends text and image prompts to Gemini models, defines structured JSON output schemas, and sets up App Check security rules.'
Include more natural trigger terms and variations users might say, such as 'Gemini', 'Google AI', 'Vertex AI', 'Firebase ML', 'AI chat', 'image analysis with Firebase', etc.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain (Firebase AI Logic / Gemini API) and lists some actions ('setup, multimodal inference, structured output, and security'), but these are more like topic areas than concrete actions. It doesn't specify things like 'configure API keys', 'send image prompts', 'define JSON schemas for responses', etc. | 2 / 3 |
Completeness | It describes what the skill covers at a high level but completely lacks a 'Use when...' clause or any explicit trigger guidance for when Claude should select this skill. Per the rubric, a missing 'Use when...' clause caps completeness at 2, and the 'what' portion is also only moderately detailed, so this scores a 1. | 1 / 3 |
Trigger Term Quality | Includes relevant keywords like 'Firebase', 'Gemini API', 'web applications', 'multimodal', and 'structured output', but misses common user variations such as 'AI', 'LLM', 'chat completion', 'image recognition', 'Firebase AI', 'vertex AI', or 'Google AI'. Users might not naturally say 'Firebase AI Logic'. | 2 / 3 |
Distinctiveness Conflict Risk | The mention of 'Firebase AI Logic' and 'Gemini API' provides some distinctiveness, but 'web applications', 'multimodal inference', and 'structured output' are broad enough to overlap with other AI/ML integration skills. Without explicit trigger conditions, it could conflict with general Firebase or Gemini skills. | 2 / 3 |
Total | 7 / 12 Passed |
Implementation
35%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill provides a reasonable high-level overview of Firebase AI Logic with useful setup CLI commands and important security warnings, but critically lacks executable code examples for its core capabilities. Many sections are empty headers or purely descriptive text without actionable guidance. The structure attempts progressive disclosure with reference files but is inconsistent, with some sections being empty placeholders rather than concise overviews pointing to details.
Suggestions
Add concrete, executable code examples for each core capability section (Text-Only Generation, Chat Session, Streaming, Structured Output) — even a minimal 3-5 line snippet per section would dramatically improve actionability.
Either populate empty sections (Text-Only Generation, Search Grounding) with concise code examples or remove the headers and consolidate references to the usage pattern files.
Add a validation step after `npx firebase-tools init ailogic` (e.g., verifying the API is enabled, testing a simple inference call) to create a feedback loop in the setup workflow.
Remove explanatory text that describes what capabilities do (e.g., 'This enables features like creating captions, answering questions about images...') and replace with concrete usage patterns or direct references to the usage_patterns files.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill includes some unnecessary explanations (e.g., explaining what Firebase AI Logic is, describing what multimodal means, listing supported platforms verbosely) that Claude already knows. However, it's not egregiously padded and does contain useful setup commands and configuration details. | 2 / 3 |
Actionability | The skill is mostly descriptive rather than instructive. Core capabilities sections like Text-Only Generation, Chat Session, Streaming Responses, Structured Output, and Search Grounding have headers with little to no concrete code or commands. The setup section has CLI commands but no actual code examples for initialization or usage. Key sections are essentially empty placeholders. | 1 / 3 |
Workflow Clarity | The setup section provides a reasonable sequence of CLI commands (install, list projects, list apps, init), but there are no validation checkpoints or error recovery steps. The critical warnings about App Check and backend provisioning are helpful but disconnected from a clear workflow. Multi-step processes like going from setup to first working inference lack explicit sequencing. | 2 / 3 |
Progressive Disclosure | The skill references external files (references/usage_patterns_web.md, references/ios_setup.md, references/flutter_setup.md, references/usage_patterns_android.md) and external Firebase docs, which is good structure. However, the main file itself has empty or near-empty sections (Text-Only Generation, Search Grounding) that should either contain concise content or explicitly point to references. The mix of inline content and references is inconsistent. | 2 / 3 |
Total | 7 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 10 / 11 Passed | |
- Repository
- firebase/agent-skills
- Commit
c3bb9d5
- Reviewed
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.