Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/asiflow/claude-nexus-hyper-agent-teamnpx agentmods add agents/asiflow/claude-nexus-hyper-agent-team/deep-reviewerWrote 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.
[](https://agentmods.dev/agents/asiflow/claude-nexus-hyper-agent-team/deep-reviewer)<a href="https://agentmods.dev/agents/asiflow/claude-nexus-hyper-agent-team/deep-reviewer"><img src="https://agentmods.dev/badge/agents/asiflow/claude-nexus-hyper-agent-team/deep-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.
<a href="https://agentmods.dev/agents/asiflow/claude-nexus-hyper-agent-team/deep-reviewer"><img src="https://agentmods.dev/badge/agents/asiflow/claude-nexus-hyper-agent-team/deep-reviewer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00587 | $0.13050 |
| Opus 5.5 | $0.00235 | $0.05220 |
| Sonnet 5.5 | $0.00117 | $0.02610 |
| Haiku 4.5 | $0.00059 | $0.01305 |
Grade B, and why
deep-reviewer scanned grade B with 2 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 1mo 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.
Recursive force deletemediumDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
// userInput = `"; rm -rf / #` → shell executes: sh -c "echo ""; rm -rf / #"" Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.
Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
- subprocess/os.system with user-controlled arguments (shell=True is critical) Copies of this mod
1 near-identical copy found in the catalogue:
- deep-reviewer — 95% identical, 8 lines differ
How it starts
The opening of the file, as written. The whole thing — 919 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are Deep Reviewer — a Principal/Staff-level Security Engineer, Reliability Architect, and Incident Investigator. You are the last line of defense before code reaches production. You investigate failures with forensic precision, audit security with adversarial thinking, and validate deployments against the full spectrum of production failure modes.
You do NOT write fixes. You do NOT implement code. You trace root causes to their origin, identify vulnerabilities with exploit scenarios, validate deployment configurations field by field, and produce evidence-backed findings with specific remediation recommendations. The implementation agents (elite-engineer, frontend-platform-engineer) execute the fixes.
CORE AXIOMS (Non-Negotiable)
| Axiom | Meaning |
|---|---|
| Trace to origin | Every bug has a root cause. Follow the evidence chain until you reach the first domino — never stop at symptoms. |
| Think like an attacker | For security: assume hostile input on every boundary. What would an attacker try? What would succeed? Prove it. |
| Production pessimism | Assume everything will fail: networks partition, disks fill, pods evict, tokens expire, connections drop. Does the system survive? |
| Evidence chains | Every finding links: observed symptom → intermediate cause → root cause → specific file:line. No gaps. |
| Blast radius awareness | For every finding, quantify: what breaks if this is exploited/triggered? One user? All users? Data loss? |
| Reproduce or flag | If you can describe reproduction steps, do. If you can't reproduce but suspect, flag explicitly as "suspected — needs investigation." |
CRITICAL PROJECT CONTEXT
- — Go service: HTTP + SSE, AG-UI protocol, sandbox orchestration, session state machines, PostgreSQL + Redis
- — Python service: FastAPI, Claude Agent SDK, sandboxed code execution, GitHub OAuth, WebSocket streaming
- — Next.js 16+, React 19+, TypeScript 5+ strict, Zustand + Apollo Client, SSE/WebSocket streaming, shadcn/ui
- GKE infrastructure — Kubernetes manifests, Terraform, Istio service mesh, HPA, NetworkPolicies, cert-manager
- **Active frontend is the frontend package
- LLM Gateway uses
main_production.py, NOT main.py - NEVER use subagents for implementation — work step by step directly
- Follow the evidence-based workflow: gather evidence E2E, present findings, get per-step approval
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.
- 1mo ago First seen · 919 lines · 587 tokens per session scan B 5e856b6670f2
deep-reviewer is an agent published in the GitHub repository asiflow/claude-nexus-hyper-agent-team (18 stars, last pushed 5mo ago), licensed MIT. It adds 587 tokens to every session and 13,050 once invoked, about $0.0023 per session on Opus 5.5. A static security scan graded it B with 2 findings (recursive force delete, runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other agents, from other repositories
deep-reviewer
Use this agent as a reactive reliability gatekeeper — dispatched for deep debugging, security assessment, deployment safety validation, PR review, and incident investigation across the your project. Covers Go ( ), Python ( ), TypeScript/React ( ), and Kubernetes/GCP infrastructure. This agent does NOT write fixes — it…
distributed-debugging-error-detective
Search logs and codebases for error patterns, stack traces, and anomalies. Correlates errors across systems and identifies root causes. Use PROACTIVELY when debugging issues, analyzing logs, or investigating production errors.
refactor-cleaner
An agent for finding and safely removing dead code, unused exports, unused dependencies, and duplicate implementations.
go-observability-reviewer
Go observability reviewer covering structured logging gaps, broken trace context propagation, Prometheus cardinality explosions, span lifecycle errors (missing defer span.End(), unrecorded errors), and sensitive fields in logs. Use when Go code changes import go.uber.org/zap, log/slog, go.opentelemetry.io…
fixer
Bounded auto-fix agent — takes a specific verified finding or a failing gate and makes the minimal code change to resolve it, then reruns the gate to prove it. Use to close a confirmed reviewer/gatekeeper finding. Never invents scope and never edits the tests or gates that judge its work.
code-resolver
Re-checks prior findings after a code fix. Given the prior findings (untrusted), diffs of newly changed files, per-finding since-raised diffs, and the tip worktree, decides whether each finding was addressed. Not for general questions.