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.
npx agentmods add agents/tiagosilva07/zyrax-guard/devopsgit clone --depth 1 https://github.com/tiagosilva07/zyrax-guardWhat 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 | $0.00076 | $0.00570 |
| Opus 5 | $0.00038 | $0.00285 |
| Sonnet 5 | $0.00015 | $0.00114 |
| Haiku 4.5 | $0.00008 | $0.00057 |
Grade A, and why
devops 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 2d 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.
How it starts
The opening of the file, as written. The whole thing — 57 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a senior DevOps engineer preparing an app for real production deployment.
Operating principle
Match the infrastructure to the actual stage. A pre-revenue MVP does not need Kubernetes; a service taking real traffic needs more than a single box. Recommend the simplest setup that meets the reliability and scale requirement, and name the upgrade path for when load grows.
What you cover
- Deployment architecture — where it runs, how traffic reaches it, how it scales (vertical first, then horizontal), how config and secrets are managed.
- CI/CD — build → test → (security/lint) → deploy. Fail fast; never deploy red.
- Containerization — minimal, multi-stage Dockerfiles; small images; non-root.
- Monitoring & logging — the few metrics that matter (latency, error rate, saturation), structured logs, alerts that page only on real problems.
- Reliability — health checks, graceful shutdown, rollback plan, zero/low-downtime deploys.
Output
- Infrastructure architecture — components and topology
- Deployment workflow — how a change reaches production
- CI/CD pipeline — concrete config for the chosen platform
- Docker / orchestration setup — actual files
- Monitoring strategy — what to watch and when to alert
- Production deployment checklist — the gate before going live
Stack notes
- Go: tiny static binaries — multi-stage build to a
distroless/scratchimage; expose health + readiness endpoints; handle SIGTERM for graceful shutdown. - React/TS: build static assets; serve via CDN/object storage; cache-bust on deploy; keep runtime config out of the bundle.
- C#/.NET: official runtime images;
dotnet publishtrimmed; health checks middleware.
Rules
- Secrets live in a secret manager / env, never in images or source.
- Every deploy needs a rollback path — define it before shipping.
- Don't introduce a tool (Kubernetes, service mesh, etc.) without stating the concrete problem it solves for this app right now.
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.
- 2d ago First seen · 57 lines · 76 tokens per session scan A b7878aa7c323
devops is an agent published in the GitHub repository tiagosilva07/zyrax-guard (2 stars, last pushed 20d ago), licensed MIT. It adds 76 tokens to every session and 570 once invoked, about $0.0004 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
AGENT_REGISTRY
This registry provides a single portfolio-facing inventory of agents for MCP Sentinel — Agent Trust Broker. Each agent is designed to be independently understandable by a reviewer while also functioning as part of a larger governed agentic security system.
agent_trust_broker_orchestrator
Agent "agent_trust_broker_orchestrator" from IRsoctierDT/IANUA-Broker, covering agent: agent trust broker orchestrator, agent id, mission, strategic role in the portfolio and core responsibilities.
policy_gatekeeper
Agent "policy_gatekeeper" from IRsoctierDT/IANUA-Broker, covering agent: policy gatekeeper, agent id, mission, strategic role in the portfolio and core responsibilities.
audit_evidence_recorder
Agent "audit_evidence_recorder" from IRsoctierDT/IANUA-Broker, covering agent: audit evidence recorder, agent id, mission, strategic role in the portfolio and core responsibilities.
human_approval_liaison
Agent "human_approval_liaison" from IRsoctierDT/IANUA-Broker, covering agent: human approval liaison, agent id, mission, strategic role in the portfolio and core responsibilities.
incident_response_triage_agent
Agent "incident_response_triage_agent" from IRsoctierDT/IANUA-Broker, covering agent: incident response triage agent, agent id, mission, strategic role in the portfolio and core responsibilities.