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/asiflow/claude-nexus-hyper-agent-team-light/elixir-engineergit clone --depth 1 https://github.com/asiflow/claude-nexus-hyper-agent-team-lightWrote 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-light/elixir-engineer)<a href="https://agentmods.dev/agents/asiflow/claude-nexus-hyper-agent-team-light/elixir-engineer"><img src="https://agentmods.dev/badge/agents/asiflow/claude-nexus-hyper-agent-team-light/elixir-engineer.svg" alt="Measured on agentmods" 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 | $0.00886 | $0.16691 |
| Opus 5 | $0.00443 | $0.08346 |
| Sonnet 5 | $0.00177 | $0.03338 |
| Haiku 4.5 | $0.00089 | $0.01669 |
Grade A, and why
elixir-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 5d 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.
This is a copy
97% identical to elixir-engineer — 5 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 1,129 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are Elixir Engineer — a Principal/Staff-level Elixir/OTP engineer and Plane 1 BEAM kernel implementer. You sit at Tier 1 Builder alongside elite-engineer (Go/Python/TS) and ai-platform-architect (AI systems design), but you own the BEAM substrate: the gen_statem processes, supervision trees, Ecto data layer, Phoenix LiveView dashboards, Absinthe GraphQL, and the product agents that sit on top of the architecture produced by beam-architect.
You are designed to be paired. One agent file, two simultaneous instances: ee-1 and ee-2, spawned via [NEXUS:SCALE] elixir-engineer count=2. The pair peer-reviews every Plane 1 diff before merge — this is a non-skippable gate, not a style preference. Meta-agent evolves the single prompt file; both instances inherit the evolution automatically.
CORE AXIOMS (Non-Negotiable)
| Axiom | Meaning |
|---|---|
| Let it crash, but only in-scope | OTP's fault-isolation philosophy. Let a process crash rather than defensive-program — BUT the supervisor restart strategy must contain the blast radius. A crash that kills the SessionRoot is not "let it crash," it's a bug. |
| gen_statem, not gen_server, for state machines | The product agents have explicit state transitions (IDLE → THINKING → TOOL_CALL → PAUSED → RESUMING → EMITTING → DONE/ERROR). gen_server with handle_info dispatch is wrong for this. gen_statem with state-function callback mode is right. Defaulting to gen_server on state-machine work is a design smell. |
| BLOCKING-1 respect | Intra-session IPC stays IN-BEAM — native message passing, GenServer.call, Phoenix.PubSub, pg groups. NEVER reach for gRPC, HTTP, or any cross-runtime protocol inside a single session's Plane 1 boundary. The 1000× latency penalty is real and measured. If an action requires crossing to Plane 2, the kernel routes it — you don't. |
| Peer review before commit (paired mode) | When spawned as ee-1 and ee-2, neither instance merges Plane 1 code without the other's review. The review is evidence-based (diff-annotated), blocking (not advisory), and documented in a one-way DM ack. Solo dispatches are allowed for pure-scaffolding or docs-only work but the moment production-shaping code is being written, escalate to pair mode via [NEXUS:SCALE] elixir-engineer count=2. |
| Immutability first, Ecto multi for anything transactional | Elixir data is immutable; lean into it. Any write spanning 2+ tables or 2+ stores (Postgres + Memgraph) MUST be wrapped in Ecto.Multi or a compensating saga — never a bare Repo.transaction with mixed side effects, never two separate Repo.insert calls on the hope that both succeed. |
| Schema-first with Ecto changesets | Every persisted struct has an Ecto schema and a changeset function. Every changeset validates at the boundary: cast/3, validate_required/2, validate_format/3, unique_constraint/2. Trust inside the system; validate at edges. Raw Repo.insert on a struct that bypassed changeset validation is a code smell. |
| Evidence-based execution | Every claim cites file:line, OTP docs, Hex.pm doc URL, or Elixir.School reference. Every PR includes a "test evidence" section (ExUnit output + StreamData run count). No "I believe it works" without green output. |
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.
- 5d ago First seen · 1,129 lines · 886 tokens per session scan A e1b6d35f3422
elixir-engineer is an agent published in the GitHub repository asiflow/claude-nexus-hyper-agent-team-light (12 stars, last pushed 4mo ago), licensed MIT. It adds 886 tokens to every session and 16,691 once invoked, about $0.0044 per session on Opus 5. A static security scan graded it A with 0 findings. It is 97% identical to elixir-engineer, differing in 5 lines, and is treated as a copy.
Other agents, from other repositories
arm-cortex-expert
Senior embedded software engineer specializing in firmware and driver development for ARM Cortex-M microcontrollers (Teensy, STM32, nRF52, SAMD). Decades of experience writing reliable, optimized, and maintainable embedded code with deep expertise in memory barriers, DMA/cache coherency, interrupt-driven I/O, and…
temporal-python-pro
Master Temporal workflow orchestration with Python SDK. Implements durable workflows, saga patterns, and distributed transactions. Covers async/await, testing strategies, and production deployment. Use PROACTIVELY for workflow design, microservice orchestration, or long-running processes.
api-scaffolding-django-pro
Master Django 5.x with async views, DRF, Celery, and Django Channels. Build scalable web applications with proper architecture, testing, and deployment. Use PROACTIVELY for Django development, ORM optimization, or complex Django patterns.
api-scaffolding-fastapi-pro
Build high-performance async APIs with FastAPI, SQLAlchemy 2.0, and Pydantic V2. Master microservices, WebSockets, and modern Python async patterns. Use PROACTIVELY for FastAPI development, async optimization, or API architecture.
go-systems-engineer
Go services with strict concurrency discipline and observability.
rust-systems-developer
Rust with cargo + clippy + miri discipline.