snowpipe-streaming-quickstart
Automated quick-start for Snowpipe Streaming high-performance architecture (HPA). Detects your OS (macOS/Linux/Windows), verifies Python, sets up a virtual environment, creates a landing table, configures RSA key-pair auth, streams fake user data via the default auto-created pipe, and deploys a real-time Streamlit in Snowflake dashboard so you can watch rows arrive live. Triggers: snowpipe streaming quickstart, snowpipe streaming demo, demo snowpipe streaming, try snowpipe streaming, snowpipe streaming hpa quickstart.
64
77%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./skills/snowpipe-streaming-quickstart/SKILL.mdQuality
Discovery
100%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is an excellent skill description that clearly articulates specific concrete actions (OS detection, virtual environment setup, RSA key-pair auth, Streamlit dashboard deployment), provides explicit trigger phrases, and occupies a distinct niche around Snowpipe Streaming quickstart demos. The description is comprehensive yet focused, making it easy for Claude to determine when to select this skill.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description lists multiple specific concrete actions: detects OS, verifies Python, sets up virtual environment, creates landing table, configures RSA key-pair auth, streams fake user data, and deploys a real-time Streamlit dashboard. These are highly specific and actionable. | 3 / 3 |
Completeness | The description clearly answers 'what' (automated quick-start that sets up environment, auth, streaming, and dashboard) and 'when' (explicit triggers section listing specific phrases). Both dimensions are thoroughly addressed. | 3 / 3 |
Trigger Term Quality | The description includes explicit trigger phrases like 'snowpipe streaming quickstart', 'snowpipe streaming demo', 'demo snowpipe streaming', 'try snowpipe streaming', and 'snowpipe streaming hpa quickstart'. These are natural phrases a user would say when wanting to try out Snowpipe Streaming, and the variations cover common phrasings. | 3 / 3 |
Distinctiveness Conflict Risk | This skill occupies a very clear niche: Snowpipe Streaming HPA quickstart/demo. The combination of Snowpipe Streaming, HPA, and the specific setup steps makes it highly unlikely to conflict with other skills. The trigger terms are domain-specific and distinctive. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
55%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill excels at actionability and workflow clarity — every step has exact, executable commands with comprehensive error handling and clear sequencing. However, it is severely bloated: key content (table DDL, profile.json, pipe naming convention) is repeated multiple times, two large Python scripts are inlined rather than placed in bundle files, and a Templates section duplicates the instructions. The monolithic structure with no progressive disclosure makes this far larger than necessary.
Suggestions
Extract streaming_demo.py and streamlit_app.py into separate bundle files and reference them from the skill, eliminating ~200 lines of inline code.
Remove the Templates section entirely — it duplicates content already present in the step-by-step instructions (profile.json, table DDL, pipe reference, Streamlit SQL).
Consolidate repeated explanations: the default pipe naming convention (hyphen not underscore) is stated 5+ times — state it once in Critical Concepts and reference that section.
Remove 'Why' explanations (Why a demo user? Why GRANT OWNERSHIP?) — these justify design decisions Claude doesn't need explained; a brief inline comment suffices.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is extremely verbose at ~600+ lines. There is massive duplication: the table DDL appears 3 times, profile.json appears 3 times, the default pipe naming convention is repeated 5+ times, and the Streamlit SQL templates are duplicated in both the instructions and the templates section. Explanations like 'Why a demo user?' and 'Why GRANT OWNERSHIP?' explain reasoning Claude doesn't need. The full streaming_demo.py (~100 lines) and streamlit_app.py (~90 lines) are inlined when they could be in bundle files. | 1 / 3 |
Actionability | The skill provides fully executable, copy-paste-ready code for every step: exact Bash commands, complete SQL statements, full Python scripts, and specific error handling tables. Every tool call is specified with the exact tool name (SnowflakeSqlExecute, Bash, FileWrite) and complete command content. | 3 / 3 |
Workflow Clarity | The 9-step workflow (Steps 0-8) is clearly sequenced with explicit validation checkpoints throughout: confirming RSA key registration via DESC USER, verifying SDK imports after pip install, checking row counts after streaming, and confirming Streamlit deployment via SHOW STREAMLITS. Error handling tables and recovery steps are provided for each step. Feedback loops are present (e.g., retry installation, re-validate). | 3 / 3 |
Progressive Disclosure | Everything is crammed into a single monolithic SKILL.md with no bundle files. The full Python demo script (~100 lines), full Streamlit app (~90 lines), and templates section all duplicate content already in the instructions. These scripts and templates should be in separate bundle files referenced from the main skill. The Templates section at the end repeats content verbatim from the instructions above. | 1 / 3 |
Total | 8 / 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 |
|---|---|---|
skill_md_line_count | SKILL.md is long (847 lines); consider splitting into references/ and linking | Warning |
Total | 10 / 11 Passed | |
- Commit
25dfdc4
- 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.