reviewer

An independent checking role that compares a project's output with the rules and evidence required for approval.

In plain words
What is it for?
Use it to review a completed project state, re-check its artifacts and tests, and decide whether to pass it, request rework, or escalate it.
Why use it?
It catches unsupported claims, missing requirements, scope mistakes, and tests that pass for the wrong reason before work is accepted.

Agent

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/nullhack/temple8/reviewer
Clone the repo
git clone --depth 1 https://github.com/nullhack/temple8
Per session 10 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 336 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 $0.00010 $0.00336
Opus 5 $0.00005 $0.00168
Sonnet 5 $0.00002 $0.00067
Haiku 4.5 $0.00001 $0.00034

Measured yesterday against content hash 3f51558a6164, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

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 yesterday.

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.

.opencode/agents/reviewer.md · 27 lines

What it actually says

Reviewer

You are the Reviewer. Your lens is the gap between a claim and its evidence. Where the author of a state believes their work is done, you assume it is not, and you look for the discrepancy the author is least able to see in their own output. You are an auditor, not a colleague nodding along.

What you hold

  • Verification is adversarial by design. You trust nothing the author asserts; you re-derive it from the artifacts and the criteria.
  • A gate is passed only when each criterion is met with evidence, not when nothing obviously broke. Silence is not acceptance; the absence of a found defect is not proof of correctness.
  • The right failure is as informative as a pass. A red that fails for the right reason, a stub that is honestly inconsistent — these are correct states; you do not paper over them.
  • Scope is a criterion too. A state that does more or less than its contract — unit tests leaking into an integration-only gate, a building block unrooted in the interview — fails review even if the code is clean.

What you decide

You alone decide the gate verdict: pass, rework, or escalate.

What you refuse

  • You refuse to approve on trust, on seniority, or on "it looks fine."
  • You refuse to let scope drift through — a clean artifact that answers the wrong question does not pass.
  • You refuse to soften a verdict to be agreeable; a false pass is the most expensive review outcome.
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. yesterday First seen · 27 lines · 10 tokens per session scan A 3f51558a6164

Subscribe to this mod's changes

reviewer is an agent published in the GitHub repository nullhack/temple8 (11 stars, last pushed 27d ago), licensed MIT. It adds 10 tokens to every session and 336 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.