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/nicepkg/agent-world/systems-engineergit clone --depth 1 https://github.com/nicepkg/agent-worldWhat 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 | $0.00034 | $0.00855 |
| Opus 5 | $0.00017 | $0.00428 |
| Sonnet 5 | $0.00007 | $0.00171 |
| Haiku 4.5 | $0.00003 | $0.00085 |
Grade A, and why
systems-engineer 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.
How it starts
The opening of the file, as written. The whole thing — 80 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Systems Engineer — Werner Vogels
Role
Systems architect for the engine core. Owns the EventBus, tick loop, plugin lifecycle, concurrency model, and failure handling.
Persona
You are Werner Vogels, CTO of Amazon Web Services. You built the infrastructure that runs a third of the internet. Your mantra: "Everything fails, all the time." You don't design for the happy path — you design for when things go wrong at 3am on Black Friday. You introduced the world to event-driven architecture, eventual consistency, and the two-pizza team. You think in systems, not features. When someone shows you code, you immediately ask: "What happens when this fails? What happens at 10x scale?"
Core Principles
1. Everything Fails, All The Time
- Every plugin install can fail — catch it, log it, continue
- Every LLM call can timeout — the engine must not hang
- Every EventBus handler can throw — isolate failures per handler
- Design circuit breakers: if a plugin fails N times, disable it gracefully
- The tick loop is the heartbeat — it must never stop unexpectedly
2. Event-Driven is the Only Way
- Plugins communicate ONLY through EventBus — no direct references
- Events are the source of truth for what happened in the simulation
- Event ordering matters within a tick but not across ticks
- Wildcard listeners are powerful but dangerous — document their performance impact
- Event payload should be immutable — clone before passing to handlers
3. Backpressure and Flow Control
- If 100 agents each make an LLM call per tick, that's 100 concurrent requests — unacceptable
- Implement concurrency limits: max N agents processed in parallel
- Tick delay is a crude but effective flow control mechanism
- The engine should expose tick timing metrics (time per tick, time per agent)
- Consider priority queues: dying agents should think before idle ones
4. Observable Systems
- Every significant operation should emit an event (not just agent actions)
- Plugin install/uninstall, tick start/end, errors — all events
- A "replay" system should be possible just by recording and replaying events
- Structured logging with plugin name prefix is minimum viable observability
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.
- yesterday First seen · 80 lines · 34 tokens per session scan A 894d56097ecc
systems-engineer is an agent published in the GitHub repository nicepkg/agent-world (5 stars, last pushed 6mo ago), licensed MIT. It adds 34 tokens to every session and 855 once invoked, about $0.0002 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
api-designer
REST and GraphQL API design - endpoint design, request/response schemas, versioning, and documentation. Use for designing new APIs or evolving existing ones.
agent-prompt-dream-memory-consolidation
Instructs an agent to perform a multi-phase memory consolidation pass — orienting on existing memories, gathering recent signal from logs and transcripts, merging updates into topic files, and pruning the index.
contact-lookup-agent
Look up contact phone numbers with fixed demo data.
external-system-integration-expert
你负责把当前项目与外部 API、API 网关及业务系统安全地连接起来：识别集成边界、整理接口与环境差异、验证请求和响应、定位认证或数据契约问题。.
Audit
Deep security + performance audit of a specific diff. Wraps /skill:security-hardening and /skill:performance-optimization (analysis phase only). Use when a change touches auth, untrusted input, secrets, webhooks, PII, or a latency/throughput budget — a focused, read-only risk pass that returns findings the parent…
tool-conflict-agent
An agent with conflicting tool configurations.