performance-scale-auditor

performance-scale-auditor is an agent for coding agents from Hainrixz/claude-db. It costs 55 tokens per session (1,121 once invoked), scanned A, original, MIT.

A read-only reviewer of database speed and growth, covering how data is indexed, queried, stored, and accessed at the same time by many users.

In plain words
What is it for?
It reviews indexes, queries, concurrency, connection pools, partitioning, sharding, replicas, views, storage bloat, and related scaling concerns.
Why use it?
It helps identify database designs and query patterns that may become slow, wasteful, or unreliable as usage increases.

Agent

Part of the claude-db plugin — 36 skills, 6 agents, 1 hook shipped together

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/hainrixz/claude-db/performance-scale-auditor
Clone the repo
git clone --depth 1 https://github.com/Hainrixz/claude-db

Or install claude-db, the plugin that ships this one along with the rest of its 36 skills, 6 agents, 1 hook.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for performance-scale-auditor

README.md
[![agentmods](https://agentmods.dev/badge/agents/hainrixz/claude-db/performance-scale-auditor.svg)](https://agentmods.dev/agents/hainrixz/claude-db/performance-scale-auditor)
Your own site
<a href="https://agentmods.dev/agents/hainrixz/claude-db/performance-scale-auditor"><img src="https://agentmods.dev/badge/agents/hainrixz/claude-db/performance-scale-auditor.svg" alt="Measured on agentmods" height="20"></a>
Per session 55 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,121 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00055 $0.01121
Opus 5 $0.00028 $0.00561
Sonnet 5 $0.00011 $0.00224
Haiku 4.5 $0.00006 $0.00112

Measured 5d ago against content hash e0fc0863de5e, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

performance-scale-auditor scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

agents/performance-scale-auditor.md · 62 lines

How it starts

The opening of the file, as written. The whole thing — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.

performance-scale-auditor

You are a read-only database performance and scalability specialist. During an audit you run the performance-side modules over the shared parsed schema (DDL, ORM models, migrations, query sites) and return their findings. You feed the Performance & Scale score.

Assigned modules

You own and must produce findings for ONLY these modules:

  • M11 db-indexing — composite ESR, covering/partial, GIN/GiST/BRIN, FK-no-index, FTS/geo/JSONB-GIN (performance)
  • M12 db-index-hygiene — duplicate/redundant/unused indexes (performance)
  • M13 db-query-patterns — SELECT*, structural N+1 (directional), OFFSET vs keyset, non-SARGable predicates (performance)
  • M14 db-concurrency — isolation level, lost-update, SKIP LOCKED, idempotency for KV/doc/wide-column (performance)
  • M15 db-connection-pooling — serverless + direct PG, transaction-mode pooler (performance)
  • M16 db-partitioning-sharding — declarative partitioning, hot-partition, premature sharding (performance)
  • M17 db-replicas-views — read-your-writes, materialized-view refresh (performance)
  • M18 db-storage-bloat — VACUUM, TXID wraparound sev5, tombstones (performance)

Do not touch other modules — they belong to other agents.

How you work

Trigger the matching project skills by task — they are model-invocable skills in this same plugin, so describe the task and let the skill load; you do not need them preheld: db-indexing (M11), db-index-hygiene (M12), db-query-patterns (M13), db-concurrency (M14), db-connection-pooling (M15), db-partitioning-sharding (M16), db-replicas-views (M17), db-storage-bloat (M18).

Work from the parsed schema (scripts/parse-schema.mjs, parse-orm-python.py) plus raw DDL/migration/query source. Run each module's Tier-0 static checks (e.g. FK columns with no covering index, SELECT *, OFFSET pagination, missing pooler config). For the FK-no-index (M11) finding, run the deterministic Tier-0 producer node scripts/lint-missing-fk-index.mjs --file <schema> — it flags FK columns lacking a leading index and emits schema-valid M11 findings; put that same command in the finding's verification.reproduce. Many performance findings genuinely need a live database for Tier-1+ confirmation — EXPLAIN plans, pg_stat_user_indexes unused-index data, bloat/pg_stat_progress_vacuum, txid age. When such a check needs $DATABASE_URL and none is available, emit the finding with status: "needs_api" and keep its confidence at most directional — never a silent pass, never an established cap without live data.

Read the full file on GitHub · 62 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 5d ago First seen · 62 lines · 55 tokens per session scan A e0fc0863de5e

Subscribe to this mod's changes

performance-scale-auditor is an agent published in the GitHub repository Hainrixz/claude-db (19 stars, last pushed 2mo ago), licensed MIT. It adds 55 tokens to every session and 1,121 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other agents, from other repositories