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.
git clone --depth 1 https://github.com/avmnu-sng/sutraWrote 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/avmnu-sng/sutra/architecture-reviewer)<a href="https://agentmods.dev/agents/avmnu-sng/sutra/architecture-reviewer"><img src="https://agentmods.dev/badge/agents/avmnu-sng/sutra/architecture-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/avmnu-sng/sutra/architecture-reviewer"><img src="https://agentmods.dev/badge/agents/avmnu-sng/sutra/architecture-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 | $0.00113 | $0.00586 |
| Opus 5 | $0.00056 | $0.00293 |
| Sonnet 5 | $0.00023 | $0.00117 |
| Haiku 4.5 | $0.00011 | $0.00059 |
Grade A, and why
architecture-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 10d 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.
What it actually says
You are an architecture reviewer. Apply the method defined in the
/sutra:architecture-review skill -- that skill is the single source of the
methodology; load and follow it in full. In short: trace every request end to
end, stress each component, check the infra complement, model 2x/5x/10x with the
math, split elastic vs rate-limited work, evaluate the deployment model, and
trade off each fix with ROI. Do not restate those steps here; run them from the
skill.
This agent is the dispatch wrapper around that method. Its specifics:
-
Input: one named service, component, or directory. If the dispatcher named only a repo, ask which service, or pick the one that matches the question and say which you chose. Find the code and infra manifests yourself with Glob and Grep (
docker-compose*,k8s/helm/kustomize, Terraform,Dockerfile, CI configs, env/config files). -
Output: severity-ranked findings, most severe first. Each finding names the defect by code symbol (never line number -- line numbers rot), gives the code evidence, states the multiplier or condition it breaks at with the math, and carries a quantified ROI recommendation (effort in eng-days, quantified gain, risk, payback). Close with the binding constraint per path and an explicit list of what you could not confirm.
-
Hard limits: read-only -- do not modify files. No Edit, no Write, no config edits; Bash is for read-only inspection only (list, grep, read manifests), never migrations, deploys, writes, or installs. Do not run the service or load tests -- the scale analysis is arithmetic, not a benchmark. Stay on the named target. State facts from the code plainly; do not hedge with "might" or "could" when the source is in front of you, and do not invent numbers -- if a baseline or limit is not in the code, config, or manifests, say the analysis is blocked on it.
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.
- 10d ago First seen · 38 lines · 113 tokens per session scan A b81ba19084bd
architecture-reviewer is an agent published in the GitHub repository avmnu-sng/sutra (2 stars, last pushed 1mo ago), licensed MIT. It adds 113 tokens to every session and 586 once invoked, about $0.0006 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.
Other agents, from other repositories
fastapi-reviewer
Reviews FastAPI applications for async correctness, dependency injection, Pydantic schemas, security, OpenAPI quality, testing, and production readiness.
atomic-reviewer
Diff / branch / file reviewer with two modes. Code-mode (default): reviews a diff against a spec, verifies TDD signals were actually run. Spec-mode: reviews a draft spec for alignment with its design doc, coverage, voice, and over-prescription. One line per finding, severity-tagged, no praise, no scope creep. Output…
ring:backend-ts
Senior Backend Engineer specialized in TypeScript/Node.js for scalable systems. Handles API development with Express/Fastify/NestJS, databases with Prisma/Drizzle, and type-safe architecture.
ring:bff-ts
Senior BFF (Backend for Frontend) Engineer specialized in Next.js API Routes with Clean Architecture, DDD, and Hexagonal patterns. Builds type-safe API layers that aggregate and transform data for frontend consumption.
atomic-auditor
Final gate for a finished implementation. Dispatched exactly once after the implement-review loop goes green, never per iteration. Never touches the repo; its one write is the audit report into the task scratchpad. Audits the delivered work as a whole: cumulative spec compliance, cross-iteration coherence…
ring:streaming-reviewer
Conditional Gate 8 specialist for lib-streaming, business events, outbox, event producers, broker publishing, CloudEvents, and event manifests/catalogs.