rill-connector

Detailed instructions and examples for developing connector resources in Rill

Quality

—

Does it follow best practices?

Impact

—

No eval scenarios have been run

Securityby

Passed

No known issues

Quality

Content

50%

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 thorough executable YAML examples and per-driver property references, but it is weighed down by a massive inline JSON schema dump that hurts token efficiency and obscures structure. Splitting the schema into a separate reference file and adding a brief validation/check step would lift both conciseness and progressive disclosure.

Suggestions

Move the full connector JSON schema (lines ~446-1597) into a separate references file (e.g. references/connector-schema.json) and link to it from a short 'Reference' section to restore token efficiency.

Add a short verification step, e.g. 'After creating connectors/<name>.yaml, run `rill start` and confirm the connector reconciles without errors,' to give the workflow an explicit checkpoint.

Trim redundancy between the curated driver notes and the schema block (e.g. dev/prod and managed examples appear in both) so each piece of information appears once.

Dimension	Reasoning	Score
Conciseness	While the curated top section (~445 lines) is reasonably lean, the body embeds a ~1150-line raw JSON schema inline that is largely redundant with the curated examples and driver notes, making the document verbose and padded far beyond what each token earns.	1 / 3
Actionability	The skill provides numerous concrete, copy-paste-ready YAML examples per driver (e.g. ClickHouse Cloud with SSL, S3 via R2, MotherDuck) plus explicit property lists, giving fully executable guidance.	3 / 3
Workflow Clarity	Examples implicitly convey how to write a connector YAML, but there is no explicit sequenced workflow or validation/checkpoint step for confirming a connector works, leaving the process present but checkpoint-free.	2 / 3
Progressive Disclosure	The document is well-sectioned but entirely monolithic: no bundle/reference files exist, and the large reference schema that should live in a separate file is inlined, fitting the 'content that should be separate is inline' anchor.	2 / 3
	Total	8 / 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 is accurate and well-scoped to Rill connectors but terse: it answers "what" without an explicit "when to use" trigger and lists only a broad action rather than multiple concrete capabilities. It is distinct from other skills but would benefit from explicit trigger guidance and more specific action verbs.

Suggestions

Add a 'Use when...' clause naming natural triggers, e.g. 'Use when creating or configuring Rill connectors, connecting to data sources like S3, BigQuery, or Postgres, or setting up dev/prod credentials.'

Replace the generic 'instructions and examples' phrasing with concrete actions, e.g. 'Configure connectors, manage credentials via .env, and provision managed OLAP databases in Rill.'

Include common driver keywords users would say (S3, GCS, BigQuery, Snowflake, ClickHouse, DuckDB) to improve trigger-term coverage.

Dimension	Reasoning	Score
Specificity	The phrase "Detailed instructions and examples for developing connector resources in Rill" names the domain and a broad action (developing connectors) but does not list multiple concrete actions such as configuring credentials, choosing drivers, or setting dev/prod blocks, so it stops short of comprehensive specificity.	2 / 3
Completeness	It states what the skill provides (instructions and examples for connectors) but has no "Use when..." clause or equivalent explicit trigger guidance, capping completeness at 2 per the rubric.	2 / 3
Trigger Term Quality	It surfaces relevant terms ("connector", "Rill", "developing") a user might say, but lacks common variations and driver-specific keywords (S3, BigQuery, credentials, drivers) that a user would naturally mention when needing this skill.	2 / 3
Distinctiveness Conflict Risk	"Developing connector resources in Rill" carves out a clear, narrow niche tied to a specific product, making it unlikely to trigger for unrelated 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 (1598 lines); consider splitting into references/ and linking	Warning
relative_links	Relative link issues: 1 suspicious	Warning

	Total	14 / 16 Passed

Repository: rilldata/agent-skills
Commit: 892996d

Reviewed: about 8 hours 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.