CtrlK
BlogDocsLog inGet started
Tessl Logo

build-paths-intermediate

Scaffold a Grok-4 tool-calling agent over an Oracle schema using langchain-oracledb + oracle-database-mcp-server + in-DB ONNX embeddings (registered MiniLM model, no external embedding API) + Open WebUI. For users who've built RAG before and want to rebuild it on the production-feeling Oracle stack.

68

Quality

83%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

88%Weight 40%Scale 1-5

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

A highly actionable, well-sequenced scaffolding procedure with concrete commands, executable code, explicit validation checkpoints, and recovery loops. Its only weaknesses are minor conciseness padding (internal friction IDs) and a progressive-disclosure design that, while well-signaled, inlines a lot of detail and rests on references that are not bundled here.

DimensionReasoningScore

Conciseness

Dense and mostly load-bearing — concrete commands, SQL, version-pinned deps, and a copy-paste agent.py — with only minor trimmable padding such as internal friction IDs ('v3 friction P0-V3-N4') and a few rationale sentences, so it is efficient but not perfectly lean.

4 / 5

Actionability

Fully executable guidance throughout: exact CLI commands (onnx2oracle load ... --name MY_MINILM_V1 --dsn ...), ready-to-run SQL smoke checks, a complete copy-paste 2-step agent.py pipeline, and precise file paths/deps covering the common per-idea cases.

5 / 5

Workflow Clarity

Clear sequenced procedure (Steps 0–5 with 3a/3b/3c sub-steps) plus explicit validation checkpoints ('Refuse if target_dir is non-empty', 'Block until OK', expected 'verify: OK'), a feedback loop (shared/verify.md recovery, max 3 retries), and a Stop conditions checklist for error recovery.

5 / 5

Progressive Disclosure

Well-sectioned body with clearly signaled, one-level-deep references to specialized materials (shared/references, shared/snippets, shared/templates, building-block skills), but it inlines substantial detail (full agent.py code, the resolve-choices table, per-idea seeding) and no bundle files are present to confirm the referenced paths resolve, so it is not the lean overview-with-delegation anchor 5.

4 / 5

Total

18

/

20

Passed

Description

78%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A specific, well-scoped description that names a concrete scaffolding task and a precise Oracle/Grok-4 stack with an explicit audience/trigger clause. It is strong on distinctiveness and solid on specificity, trigger terms, and completeness, though the 'when' is framed as audience rather than a direct 'Use when...' trigger.

Suggestions

Add an explicit 'Use when ...' trigger clause naming the natural phrases users would say (e.g. 'Use when the user wants to build a RAG or NL2SQL agent on Oracle, calls for tool-calling over a live schema, or asks for in-database ONNX embeddings') to lift completeness toward 5.

Frame the capability as multiple distinct actions (scaffold the agent, register the in-DB ONNX embedder, wire the MCP SQL server, expose it via Open WebUI) rather than one 'Scaffold' verb plus a component list, to reach the comprehensive-action anchor 5.

Include a few synonym/variation trigger terms (e.g. 'NL2SQL', 'text-to-SQL', 'Oracle AI Vector Search', 'OCI GenAI') alongside 'RAG' and 'Oracle' for broader keyword coverage.

DimensionReasoningScore

Specificity

Names one concrete action ('Scaffold a Grok-4 tool-calling agent over an Oracle schema') plus a specific component stack (langchain-oracledb, oracle-database-mcp-server, in-DB ONNX, Open WebUI), but the action is singular rather than a list of distinct actions, so it stops short of the comprehensive multi-action anchor 5.

4 / 5

Completeness

Clearly states 'what' (scaffold an agent over an Oracle schema with a named stack) and gives equivalent trigger guidance ('For users who've built RAG before and want to rebuild it on the production-feeling Oracle stack'), but the 'when' is audience-framed rather than an explicit 'Use when...' trigger phrase, so it is not the fully explicit anchor 5.

4 / 5

Trigger Term Quality

Good natural-term coverage ('RAG', 'Oracle', 'tool-calling agent', 'Grok-4', 'Open WebUI') that a user would plausibly say, but lacks synonyms/common variations needed for the comprehensive anchor 5.

4 / 5

Distinctiveness Conflict Risk

A clear niche — Grok-4 + Oracle schema + in-DB ONNX + oracle-database-mcp-server + Open WebUI — with distinct, highly specific triggers and minimal overlap risk with other skills.

5 / 5

Total

17

/

20

Passed

Validation

87%

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

Validation14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

referenced_paths_exist

Referenced path issues: 3 missing

Warning

Total

14

/

16

Passed

Repository
oracle-devrel/oracle-ai-developer-hub
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.