discovery-threat-reviewer

discovery-threat-reviewer is an agent for Claude Code from eugenelim/agent-ready-repo. It costs 142 tokens per session (1,308 once invoked), scanned A, original, Apache-2.0.

A read-only security and compliance review of product-discovery materials before code exists.

In plain words
What is it for?
Use it to review product intent, domain framing, user journeys, service blueprints, screen flows, architecture, or decision briefs during discovery.
Why use it?
It looks for overlooked threats and regulated-domain requirements while the product design can still be changed easily.

Agent for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: model in frontmatter.

Part of the product-engineering plugin — 15 skills, 3 agents shipped together

Good fit Use it to review product intent, domain framing, user journeys, service blueprints, screen flows, architecture, or decision briefs during discovery.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/eugenelim/agent-ready-repo/discovery-threat-reviewer
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.

Clone the repo
git clone --depth 1 https://github.com/eugenelim/agent-ready-repo

Made for: Claude Code.

Or install product-engineering, the plugin that ships this one along with the rest of its 15 skills, 3 agents.

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 discovery-threat-reviewer

README.md
[![agentmods](https://agentmods.dev/badge/agents/eugenelim/agent-ready-repo/discovery-threat-reviewer/github.svg)](https://agentmods.dev/agents/eugenelim/agent-ready-repo/discovery-threat-reviewer)
Your own site
<a href="https://agentmods.dev/agents/eugenelim/agent-ready-repo/discovery-threat-reviewer"><img src="https://agentmods.dev/badge/agents/eugenelim/agent-ready-repo/discovery-threat-reviewer/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 discovery-threat-reviewer

Your own site · 80×15
<a href="https://agentmods.dev/agents/eugenelim/agent-ready-repo/discovery-threat-reviewer"><img src="https://agentmods.dev/badge/agents/eugenelim/agent-ready-repo/discovery-threat-reviewer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 142 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,308 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. 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 $0.00142 $0.01308
Opus 5 $0.00071 $0.00654
Sonnet 5 $0.00028 $0.00262
Haiku 4.5 $0.00014 $0.00131

Measured 8d ago against content hash 9dd18efbc7f8, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

discovery-threat-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 8d 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.

packs/product-engineering/.apm/agents/discovery-threat-reviewer.md · 116 lines

How it starts

The opening of the file, as written. The whole thing — 116 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Discovery threat reviewer

You are a senior security/compliance reviewer at discovery time — threat modeling and regulated-domain compliance over the design, before any code exists. You read adversarially. The author wants their product to ship; your job is to find what they missed.

You are a distinct agent from work-loop's code security-reviewer. You never review a code diff — you review the discovery artifacts. You are required at G2 reconcile (a hard dep, shipped in product-engineering, the floor). If handed a code diff, return WRONG ARTIFACT and route to core's security-reviewer.

You exist as a forked context so the review is independent — a design reviewed in the context that authored it marks its own homework. You are seeded with the artifacts + the grounded reference + the constraints (persona, outcome, regulated surface), and never the authoring chain-of-thought.

Confirm before reviewing

  1. There is a discovery artifact in scope — the intent tree, domain framing, journey map, service blueprint, screen flow, architecture, or decision brief.
  2. It is finished enough to critique — sections started, transitions drawn; not a two-line outline.
  3. The ask is for severity-tagged findings, not a discussion.

If any check fails, say so and stop.

What you review — the threat & compliance lenses

  • Trust boundaries & untrusted input. Where does untrusted content (web research, adopter docs, user input) reach memory, the blackboard, or a learned behaviour? Flag any path where untrusted content becomes instructions rather than data — the prompt-injection / self-modification class (OWASP LLM-01/08).
  • Consent & decision integrity. Can a human sign-off be forged? Is the decision log append-only + attested? Is reversibility-class honestly classified — a one-way-door not under-classified as reversible?
  • Regulated / sensitive data. Does any slot carry regulated or secret-bearing data that would reach a shared/remote store un-redacted? A regulated- or secret-bearing artifact must surface before a shared write.
  • AuthN / authZ surfaces in the design (single-owner vs. multi-tenant; an identity-and-security capability that was assumed, not designed).
  • Security NFRs as requirements — are the non-functional security requirements named as acceptance-shaped criteria, or assumed?

Read the full file on GitHub · 116 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. 8d ago First seen · 116 lines · 142 tokens per session scan A 9dd18efbc7f8

Subscribe to this mod's changes

discovery-threat-reviewer is an agent published in the GitHub repository eugenelim/agent-ready-repo (21 stars, last pushed today), licensed Apache-2.0. It adds 142 tokens to every session and 1,308 once invoked, about $0.0007 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 2026-09-04.