postgres-query

Run PostgreSQL queries for testing, debugging, and performance analysis. Use when you need to query the database directly, run EXPLAIN ANALYZE, compare query results, or test SQL optimizations. Always pass a target — `--prod` or `--dev` — because prod and dev are different databases and the default is prod. Always uses read-only connections unless explicitly directed otherwise.

Quality

82%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

77%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 body with excellent executable examples and strong validation checkpoints for destructive/database operations. It is held back by verbosity — dated incident narratives and a history-correction block that bloat the token budget — and by inlined setup detail that a references file could absorb.

Suggestions

Move the dated DATABASE_URL incident narrative and the DDL ownership-history correction into a short 'Known pitfalls / past incidents' (or deprecated-claim) section so the main flow stays lean; lead each with the current rule, not the old wrong one.

Pull the SSH-tunnel setup recipes for --dev and --notifications (the 'Ask an infra owner…' boilerplate and the DEV_DATABASE_URL/NOTIFICATION_DB_REPLICA_URL wiring steps) into a references file (e.g. references/tunnels.md) and link to it from a one-line pointer.

Tighten the 'There is a second skill with this name' warning and the write-guard caveat to the single load-bearing sentence each, cutting the redundant restatement of the banner behavior.

Dimension	Reasoning	Score
Conciseness	Mostly efficient and free of basic-concept padding, but the long DATABASE_URL provenance narrative ("That cost an hour and produced a confidently wrong root cause on 2026-08-16…") and the multi-paragraph history-correction block on DDL push it past a lean budget. The dated incident anecdotes are precisely the time-sensitive material the rubric says should be trimmed unless quarantined in a deprecation section.	3 / 5
Actionability	Every guidance is copy-paste-ready: exact `node .claude/skills/postgres-query/query.mjs --prod "…"` invocations, a full target/options flag table, and worked examples covering simple, dev, EXPLAIN, timeout, file, JSON, and notifications cases. Fully executable with no pseudocode.	5 / 5
Workflow Clarity	Sequences are explicit and paired with validation: the mandatory per-run banner ("Read that line before you trust a result"), the read-only default plus `--writable` permission gate, the replica-rejects-writable error rule, and the performance-comparison recipe (same target on both runs). Destructive/batch operations carry clear feedback loops and warnings, hitting the top anchor.	5 / 5
Progressive Disclosure	No bundle files exist (references/, scripts/, assets/ are absent) and the skill is a single 228-line SKILL.md with the one script invoked by path — appropriate for a small skill, but the dev/notifications SSH-tunnel setup and the DATABASE_URL provenance detail are inlined material that could live one level deep. Structure is reasonable with clear section headers, so it sits above the unstructured 2 anchor but not at the cleanly-split 4.	3 / 5
	Total	16 / 20 Passed

Description

87%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 strong, third-person description that explicitly states both what the skill does and when to use it, with concrete trigger phrases and a niche-clear scope. It loses a little on specificity (a few actions live only in the body) and on trigger-term breadth (missing the "postgres"/synonym vocabulary users often say).

Dimension	Reasoning	Score
Specificity	Lists several concrete actions — "run EXPLAIN ANALYZE", "compare query results", "test SQL optimizations" — alongside the core query action, with only minor coverage gaps (e.g. DDL/migrations, index verification). It does not reach the comprehensive 5-anchor bar of many distinct concrete actions.	4 / 5
Completeness	Clearly answers both "what" ("Run PostgreSQL queries for testing, debugging, and performance analysis") and "when" ("Use when you need to query the database directly, run EXPLAIN ANALYZE, compare query results, or test SQL optimizations") with concrete trigger phrases, matching the top anchor.	5 / 5
Trigger Term Quality	Includes natural phrases a user would say ("query the database directly", "compare query results", "test SQL optimizations", "performance analysis") with good coverage, but misses common synonyms and the product name itself ("postgres", "psql", "query plans", "tune/benchmark"). That is one tier below the comprehensive-with-synonyms 5 anchor.	4 / 5
Distinctiveness Conflict Risk	Scopes to a clear niche (direct PostgreSQL querying against named prod/dev targets with read-only default) and adds a distinguishing constraint ("Always pass a target — --prod or --dev"), giving it minimal conflict risk with unrelated skills.	5 / 5
	Total	18 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: civitai/civitai
Path: .claude/skills/postgres-query/SKILL.md
Commit: 00346e6

Reviewed: 3 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.