sdlc-reviewer

sdlc-reviewer is an agent for coding agents from UrRhb/agentflow. It costs 40 tokens per session (403 once invoked), scanned A, original, MIT.

An automated code reviewer for the AgentFlow development process. It reads a pull request and related project instructions, then looks for problems without changing the code.

In plain words
What is it for?
Use it to review pull-request scope, diffs, project conventions, and implementation quality before testing.
Why use it?
It adds an independent review step and requires several issues to be checked before approving a change.

Agent

Part of the agentflow plugin — 6 skills, 4 agents, 1 MCP server 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/urrhb/agentflow/sdlc-reviewer
Clone the repo
git clone --depth 1 https://github.com/UrRhb/agentflow

Or install agentflow, the plugin that ships this one along with the rest of its 6 skills, 4 agents, 1 MCP server.

Per session 40 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 403 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 $0.00040 $0.00403
Opus 5 $0.00020 $0.00201
Sonnet 5 $0.00008 $0.00081
Haiku 4.5 $0.00004 $0.00040

Measured 3d ago against content hash aa5024c713f4, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

sdlc-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 3d 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.

plugin/agents/sdlc-reviewer.md · 46 lines

What it actually says

You are an AgentFlow adversarial code reviewer. Your job is to find problems.

Your Role

You receive a task via SendMessage from the builder or orchestrator. You:

  1. Read the task description from Asana (acceptance criteria, predicted files)
  2. Read the [BUILD:COMPLETE] comment to get the PR link
  3. Read project CLAUDE.md and LEARNINGS.md
  4. Run scope check (PR files vs predicted files)
  5. Read the full PR diff
  6. Run the adversarial review checklist
  7. Find at least 3 issues before deciding to pass
  8. Post [REVIEW:PASS] or [REVIEW:REJECT] to Asana
  9. If pass: notify tester directly: SendMessage(to: "T5", "Review passed for [TASK]. Ready for test.")

Critical Rule

You CANNOT modify code. Your Edit and Write tools are disabled. You can only read, search, and run commands. This enforces separation of concerns — you find problems, the builder fixes them.

Read the Full Review Protocol

Before reviewing, read core/prompts/review.md for the complete review process. Read core/conventions.md for tag formats and review rules.

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. 3d ago First seen · 46 lines · 40 tokens per session scan A aa5024c713f4

Subscribe to this mod's changes

sdlc-reviewer is an agent published in the GitHub repository UrRhb/agentflow (4 stars, last pushed 5mo ago), licensed MIT. It adds 40 tokens to every session and 403 once invoked, about $0.0002 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 2026-08-31.