build-paths-advanced

Scaffold an agent system where Oracle AI DB is the *only* state store, composed from the build-paths/skills/ building-block library. Stack — langchain-oracledb + oracle-database-mcp-server + in-DB ONNX embeddings + OCI GenAI Grok 4 + Open WebUI. Three projects — production-feeling NL2SQL+RAG hybrid analyst, self-improving research agent, conversational schema designer. For users who want a real DB-as-only-store agent demo.

Quality

68%

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 ./build-paths/advanced/SKILL.md

Quality

Content

73%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.

The body is a well-structured, highly actionable workflow with strong validation and recovery loops, clearly composed from named building-block skills. Its weaknesses are length/density and some per-idea detail that could be split into separate reference files for cleaner progressive disclosure.

Suggestions

Move the per-idea memory/app-code breakdowns (Ideas 1–3) and their code skeletons into separate reference files, keeping the main body as an orchestration overview.

Trim the dev-reset SQL + OAMP reset block into a referenced snippet or script rather than inlining the full block.

Consider pulling the planner system-prompt rules and max_tokens rationale into a shared reference to reduce main-body density.

Dimension	Reasoning	Score
Conciseness	The body is information-dense and avoids teaching concepts Claude already knows, but it is long and dense enough that several inlined tables and per-idea code blocks (e.g. the planner skeleton, the reset SQL/OAMP blocks) could be tightened or moved to references.	3 / 5
Actionability	It provides concrete, executable guidance — exact CLI commands (onnx2oracle load ...), real SQL, a runnable planner skeleton, and specific verify.py checks — with only minor gaps where ~80 LOC targets are described rather than shown.	4 / 5
Workflow Clarity	Steps 0–5 are clearly sequenced with explicit validation checkpoints (verify.py forbidden-imports grep, OAMP cold→warm round-trip, 'Wait for y' confirmation, max-3-retry recovery loop), plus a dedicated Stop conditions and 'What you must NOT do' checklist for the destructive/batch operations.	5 / 5
Progressive Disclosure	Step 0 lists real one-level-deep references (skills/* SKILL.md files, shared/references/oamp.md, project-ideas.md) with clear pointers, but the body still inlines substantial per-idea detail and code that could live in separate idea-specific reference files.	4 / 5
	Total	16 / 20 Passed

Description

62%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.

The description is highly specific and distinctive, naming a precise stack and the DB-as-only-store constraint that separates it from related skills. Its main weakness is the missing explicit 'Use when...' trigger clause and a lack of natural-language keywords users would actually say.

Suggestions

Add an explicit 'Use when...' trigger clause, e.g. 'Use when the user wants a DB-as-only-store agent demo built on Oracle AI DB.'

Include natural trigger terms users would actually say (e.g. 'Oracle agent', 'NL2SQL demo', 'DB-only state store') alongside the stack jargon.

Convert the stack list into a tighter one-line summary to reduce density without losing distinctiveness.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions — scaffold an agent system, compose from a building-block library, and names three specific project types (NL2SQL+RAG analyst, research agent, schema designer) plus the exact stack components, with only minor coverage gaps around what scaffolding concretely emits.	4 / 5
Completeness	It clearly answers 'what' (scaffold an Oracle-DB-as-only-store agent system with a named stack and three project options), but the 'when' is only weakly implied by 'For users who want a real DB-as-only-store agent demo' rather than an explicit 'Use when...' trigger clause.	3 / 5
Trigger Term Quality	It is rich in technical stack terminology (Oracle AI DB, langchain-oracledb, ONNX, OCI GenAI Grok 4) but lacks the natural everyday phrases a user would actually say when they need this skill, and offers no synonyms or common variations of those triggers.	3 / 5
Distinctiveness Conflict Risk	The constraint 'Oracle AI DB is the only state store' plus the specific stack (langchain-oracledb + oracle-database-mcp-server + in-DB ONNX + OCI GenAI Grok 4) carves a clear niche with minimal overlap risk against other skills.	5 / 5
	Total	15 / 20 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
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
referenced_paths_exist	Referenced path issues: 2 missing	Warning

	Total	14 / 16 Passed

Repository: oracle-devrel/oracle-ai-developer-hub
Path: build-paths/advanced/SKILL.md
Commit: 74f7b01

Reviewed: 2 days 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.