rill-connector

Detailed instructions and examples for developing connector resources in Rill

Quality

61%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/rill-connector/SKILL.md

Quality

Content

65%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is highly actionable with complete YAML examples and good structural organization, but it is bloated by an inlined full JSON schema and offers no file-level progressive disclosure or validation checkpoints. Moving the schema to a referenced file and adding a verify step would address the weakest dimensions.

Suggestions

Move the ~1150-line JSON schema dump out of SKILL.md into a referenced file (e.g. references/connector-schema.yaml) and link to it, keeping SKILL.md as a concise overview.

Add an explicit validation/reconcile checkpoint after authoring a connector (e.g. the command to apply/validate and how to read a connection failure), creating a feedback loop.

Deduplicate drivers that already appear in both "Driver-Specific Notes" and "Examples" to reduce token cost without losing clarity.

Dimension	Reasoning	Score
Conciseness	The curated first half (intro, core concepts, driver notes, examples) is lean and Rill-specific, but ~1150 lines of raw JSON schema are inlined and largely redundant with the curated driver sections and examples above, which is verbosity the rubric penalizes even when accurate.	2 / 3
Actionability	It provides numerous complete, copy-paste-ready YAML examples for every driver (e.g. the ClickHouse Cloud, S3, and Postgres connector blocks) with concrete property values, fully executable as written.	3 / 3
Workflow Clarity	There is a logical progression (naming → secrets → managed → access modes → dev/prod) and clear sections, but no explicit validation checkpoint or feedback loop for the developer to run after authoring a connector, capping this at 2.	2 / 3
Progressive Disclosure	Sections are well-organized with clear headers, but everything is inlined into a single monolithic file with no reference bundle files; the massive JSON schema is exactly "content that should be separate is inline."	2 / 3
	Total	9 / 12 Passed

Description

57%

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 names a specific, distinctive niche (Rill connectors) but is generic about capabilities and lacks an explicit "when to use" trigger. Adding concrete actions and a Use-when clause would lift the weaker dimensions.

Suggestions

Add an explicit trigger clause, e.g. "Use when creating or configuring Rill connectors to external data sources, OLAP engines, or service integrations."

Replace the content-meta phrasing with concrete actions: "Create, configure, and manage connector YAML resources in Rill across drivers like DuckDB, ClickHouse, S3, BigQuery, and Snowflake."

Include natural variations users would say ("data source", "credentials", "database connection") to improve trigger-term coverage.

Dimension	Reasoning	Score
Specificity	Quotes the domain ("developing connector resources in Rill") and a single action verb ("developing"), but "instructions and examples" describes the skill's contents rather than concrete capabilities, so it stops short of listing multiple specific actions.	2 / 3
Completeness	It answers "what" (instructions/examples for developing Rill connectors) but provides no "Use when..." clause or equivalent explicit trigger guidance, which caps completeness at 2 per the rubric.	2 / 3
Trigger Term Quality	It surfaces real natural keywords ("connector", "Rill") a user would say, but omits common variations like "data source", "credentials", "configure", or "connect to a database".	2 / 3
Distinctiveness Conflict Risk	"connector resources in Rill" is a narrow, product-specific niche unlikely to fire for unrelated skills; it is clearly distinguishable from generic data or document skills.	3 / 3
	Total	9 / 12 Passed

Validation

87%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (1660 lines); consider splitting into references/ and linking	Warning
relative_links	Relative link issues: 1 suspicious	Warning

	Total	14 / 16 Passed

Repository: rilldata/agent-skills
Path: skills/rill-connector/SKILL.md
Commit: aa23a8e

Reviewed: 1 day ago

Table of Contents

Discovery Implementation Validation

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.