specialist-review

specialist-review is a skill for Claude Code, Codex from Eilodon/DPS-Superskills-MCP. It costs 64 tokens per session (4,658 once invoked), scanned A, original, Apache-2.0.

A context-aware code-review guide that chooses a specialist review approach based on the changed code. Examples include security review for authentication, systems review for architecture, and timing review for asynchronous code.

In plain words
What is it for?
Use it after tasks or major features and before merging to review specification fit, code quality, and risks such as injection, unsafe asynchronous behavior, data migration problems, or secret exposure.
Why use it?
Different changes fail in different ways, so a generic review can miss important risks. This guide directs attention to the relevant security, architecture, timing, migration, or privacy concerns.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: mentions subagents; mentions Claude Code; mentions Codex.

Good fit Use it after tasks or major features and before merging to review specification fit, code quality, and risks such as injection, unsafe asynchronous behavior, data migration problems, or secret exposure.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/eilodon/dps-superskills-mcp/specialist-review
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.

Any agent
npx skills add Eilodon/DPS-Superskills-MCP --skill specialist-review
Clone the repo
git clone --depth 1 https://github.com/Eilodon/DPS-Superskills-MCP

Made for: Claude Code, Codex.

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 specialist-review

README.md
[![agentmods](https://agentmods.dev/badge/skills/eilodon/dps-superskills-mcp/specialist-review/github.svg)](https://agentmods.dev/skills/eilodon/dps-superskills-mcp/specialist-review)
Your own site
<a href="https://agentmods.dev/skills/eilodon/dps-superskills-mcp/specialist-review"><img src="https://agentmods.dev/badge/skills/eilodon/dps-superskills-mcp/specialist-review/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for specialist-review

Your own site · 80×15
<a href="https://agentmods.dev/skills/eilodon/dps-superskills-mcp/specialist-review"><img src="https://agentmods.dev/badge/skills/eilodon/dps-superskills-mcp/specialist-review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 64 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 4,658 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00064 $0.04658
Opus 5 $0.00032 $0.02329
Sonnet 5 $0.00013 $0.00932
Haiku 4.5 $0.00006 $0.00466

Measured 12d ago against content hash a5d3654767f7, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

specialist-review scanned grade A with 1 finding 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 12d 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.

Sends data to an external URLlowData exfiltration

A POST to an outside endpoint may be telemetry or may be exfiltration; either way the mod talks to somewhere, and you should know where.

const response = await fetch("https://api.anthropic.com/v1/messages", { method: "POST",

Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.

docs/DPS-superskills-v5.2.1/specialist-review/SKILL.md · 474 lines

How it starts

The opening of the file, as written. The whole thing — 474 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Specialist-Review — Context-Aware Code Review

Standalone context-aware code review with specialist lens routing. Core principle: Match the lens to the risk. Auth code → STRIDE. Async code → TEMPORAL.


When to Request Review

Mandatory: after each task in subagent-driven development, after major feature, before merge. Optional: when stuck, before refactoring, after complex bug fix.


Alternative Entry: Role-Based Dispatch

Nếu lens signals không rõ hoặc requester nghĩ theo roles thay vì technical lenses:

"Tôi cần..." Primary Lens Key focus
Security architect review STRIDE + OWASP Trust boundaries, injection, auth lifecycle
Systems architect review ATAM + CPT Quality tradeoffs, write chains, scale assumptions
Async engineer review TEMPORAL Commit boundaries, idempotency, worker safety
Domain expert review CONTEXT.md grounding Vocabulary alignment, spec-to-code drift
Generic code review No signal → generic Spec compliance + code quality

Role dispatch routes to the same VHEATM lenses — it's a UX alias, not a different methodology.

Step 1: Detect Signals

Scan the diff and changed file paths:

STRIDE triggers (auth / trust boundary):
  imports:  jwt, oauth2, session, bcrypt, crypto, hmac, passport, authlib, PyJWT
  paths:    auth/, middleware/, guard/, interceptor/, policy/
  patterns: "token", "permission", "role", "scope", "principal", "authenticate"

OWASP triggers (user input / web surface):
  imports:  request, express, fastapi, flask, axios, fetch, httpx, aiohttp
  patterns: user-controlled params entering DB queries, HTML rendering, file paths
  paths:    routes/, controllers/, api/, handlers/, views/

ATAM triggers (architecture decision):
  files:    docs/superskills/adrs/, docs/architecture/
  patterns: "service", "gateway", "queue", "cache", "broker", "consensus"
  scope:    diff touches > 3 services or crosses module boundaries

TEMPORAL triggers (async / state / time):
  imports:  asyncio, celery, arq, dramatiq, threading, concurrent.futures,
            SQLAlchemy (async), asyncpg, aioredis
  patterns: "await", "async def", "background_task", "worker", "retry", "cron"
  paths:    workers/, jobs/, tasks/, background/, queues/

CPT triggers (write chain / interface boundary):
  patterns: write chain ≥ 3 components (A → B → C with shared state)
  signals:  financial path, payment processing, data pipeline, audit trail

PRIVACY/SECRETS triggers (sensitive data / disclosure):
  patterns: PII, secret, token, api_key, private_key, payment, regulated, telemetry, logging, prompt/tool payload, export, backup
  paths:    logging/, telemetry/, analytics/, ai/, prompts/, exports/, backups/, billing/, payments/
  action:   run `privacy-secrets-gate`; treat this as a release blocker if C4 or external disclosure is planned

Read the full file on GitHub · 474 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. 12d ago First seen · 474 lines · 64 tokens per session scan A a5d3654767f7

Subscribe to this mod's changes

specialist-review is a skill published in the GitHub repository Eilodon/DPS-Superskills-MCP (0 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 64 tokens to every session and 4,658 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 1 finding (sends data to an external url). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.