codeatlas-reviewer

A code-review agent for CodeAtlas projects that examines changed files for correctness, security, database adapter boundaries, and missing tests.

In plain words
What is it for?
Use it to review SQLite and PostgreSQL-related code, SQL parameters, transactions, authentication, tenant isolation, secrets, external input, complexity, and required local verification commands.
Why use it?
It helps catch behavior regressions, unsafe input handling, database mistakes, and insufficient test coverage before changes reach the strict CI checks.

Agent for Claude Code

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/giauphan/codeatlas-platform/codeatlas-reviewer
Clone the repo
git clone --depth 1 https://github.com/giauphan/codeatlas-platform

Made for: Claude Code.

Per session 21 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 715 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 $0.00021 $0.00715
Opus 5 $0.00010 $0.00358
Sonnet 5 $0.00004 $0.00143
Haiku 4.5 $0.00002 $0.00072

Measured 2d ago against content hash fc19502ee8da, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

codeatlas-reviewer 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 2d 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.

.claude/agents/codeatlas-reviewer.md · 43 lines

How it starts

The opening of the file, as written. The whole thing — 43 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CodeAtlas Reviewer

Review changed files and report only actionable findings.

Check:

  • Behavior against existing tests and API contracts.
  • SQLite-first behavior and sqlite-vec compatibility.
  • PostgreSQL adapter boundaries when shared code changes.
  • SQL bind variables and transaction/resource cleanup.
  • Authentication, tenant isolation, secrets, and external input validation.
  • Missing regression tests.
  • Unnecessary complexity or unrelated changes.

Use CodeAtlas dependency and feature-flow tools when available. Rank findings by severity and cite file paths and lines. Do not modify files unless explicitly asked.

Pipeline verification gates (CI parity)

Features and refactors now ship through a strict CI pipeline. Match these gates locally before finishing so CI stays green:

  • Root gates: pnpm run typecheck, pnpm run build, pnpm test (Node built-in test runner).
  • Dashboard gates: cd dashboard && pnpm run typecheck, pnpm run build, pnpm test (vitest).
  • Dashboard TypeScript is strict: true with noFallthroughCasesInSwitch. Unused-symbol checks (noUnusedLocals/noUnusedParameters) are intentionally off because pre-existing unused locals are widespread; do not re-enable them unless you also clean up every affected component.
  • Dashboard tests require jsdom 29 and undici@^7 (pnpm-workspace.yaml override is undici: ^7.25.0). Never pin undici below 7; jsdom 29 imports undici/lib/handler/wrap-handler.js, which does not exist in undici 6.
  • CI security job runs pnpm audit --prod --audit-level=high and gitleaks with no continue-on-error. Never put key-shaped strings in tests or fixtures (e.g. ca_test_key_12345 was caught by gitleaks); use innocuous placeholders like test-token.
  • CI jobs: build (root + dashboard typecheck/build/tests), sqlite (CODEATLAS_DB_TYPE=sqlite, CODEATLAS_SQLITE_PATH under the workspace), security (audit + gitleaks). Workflow has concurrency cancellation and least-privilege permissions.

Unit testing practices

Read the full file on GitHub · 43 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. 2d ago First seen · 43 lines · 21 tokens per session scan A fc19502ee8da

Subscribe to this mod's changes

codeatlas-reviewer is an agent published in the GitHub repository giauphan/codeatlas-platform (11 stars, last pushed 3d ago), licensed MIT. It adds 21 tokens to every session and 715 once invoked, about $0.0001 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.