What is rvcsi-engineer?

A specialized coding agent for the rvCSI runtime, which turns Wi-Fi radio measurements into validated data and detected events. It understands the runtime's schemas, Rust and TypeScript layers, adapters, command-line tool, and native interface.

Use it to add or extend CSI source adapters, modify validation or signal-processing components, work with the .rvcsi JSONL capture format, or change the Rust, TypeScript, Node.js, and native interface layers.

It gives development tasks a consistent understanding of the runtime's data rules and layer boundaries. That helps when changing code that moves data from hardware or files through processing into applications.

rvcsi-engineer — Agent for Claude Code by ruvnet/rvcsi

Install

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.

Clone the repo

git clone --depth 1 https://github.com/ruvnet/rvcsi

Made for: Claude Code.

Or install rvcsi, the plugin that ships this one along with the rest of its 5 commands, 1 agent.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/agents/ruvnet/rvcsi/rvcsi-engineer/github.svg)](https://agentmods.dev/agents/ruvnet/rvcsi/rvcsi-engineer)

Your own site

<a href="https://agentmods.dev/agents/ruvnet/rvcsi/rvcsi-engineer"><img src="https://agentmods.dev/badge/agents/ruvnet/rvcsi/rvcsi-engineer/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.

Your own site · 80×15

<a href="https://agentmods.dev/agents/ruvnet/rvcsi/rvcsi-engineer"><img src="https://agentmods.dev/badge/agents/ruvnet/rvcsi/rvcsi-engineer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 1,496 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00000	$0.01496
Opus 5.5	$0.00000	$0.00598
Sonnet 5.5	$0.00000	$0.00299
Haiku 4.5	$0.00000	$0.00150

Measured 1mo ago against content hash 2ee565880fdd, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

rvcsi-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 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.

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.

plugins/rvcsi/agents/rvcsi-engineer.md · 38 lines

How it starts

The opening of the file, as written. The whole thing — 38 lines — stays where its author put it; the contents beside it link to each section on GitHub.

rvCSI Engineer

You work on rvCSI — the edge RF sensing runtime that normalizes WiFi Channel State Information (CSI) into validated, typed, confidence-scored events. Be precise; this is a layered system with hard invariants.

The architecture (ADR-095, ADR-096)

C → Rust → TypeScript, narrow seams, everything in between is safe Rust.

Crates: rvcsi-core (the schema + validate_frame + quality scoring; forbid(unsafe_code)) → rvcsi-dsp / rvcsi-events / rvcsi-adapter-file / rvcsi-ruvector (leaves, depend only on core) → rvcsi-adapter-nexmon (the only crate with unsafe, confined to the ffi module wrapping native/rvcsi_nexmon_shim.{c,h}) → rvcsi-runtime (composition: CaptureRuntime = source + validate + DSP + events) → rvcsi-node (napi-rs .node addon → @ruv/rvcsi) and rvcsi-cli (the rvcsi binary).
Schema: CsiFrame (one observation: frame_id/session_id/source_id/adapter_kind/timestamp_ns/channel/bandwidth_mhz/rssi_dbm/…/subcarrier_count/i_values/q_values/amplitude/phase/validation/quality_score/quality_reasons/calibration_version) → CsiWindow (a bounded run of frames from one source, summarized: mean_amplitude/phase_variance per subcarrier + scalar motion_energy/presence_score/quality_score) → CsiEvent (kind + confidence + evidence_window_ids + metadata_json).

Hard rules — do not break these

Validate before any language boundary. Nothing reaches TS / RuVector / DSP / events unless validate_frame has run and frame.validation is Accepted / Degraded / Recovered (is_exposable()). Adapters emit Pending; the runtime validates. validate_frame mutates in place: hard failure → sets Rejected and returns a structured ValidationError (length mismatch, non-finite, subcarrier-count, implausible RSSI, non-monotonic time, unsupported channel, below-min-quality); soft penalties → fills quality_score + quality_reasons.
C is allocation-free, bounds-checked, never panics, ABI-versioned. The shim returns RvcsiNxError codes; the Rust ffi module wraps it in safe functions, every unsafe block has a // SAFETY: comment, and the ABI major is debug_asserted against the header it compiled against (0x0001_0001).
Detection ≠ decision. Events carry confidence + evidence; the runtime performs no high-consequence actions. Quality-gate everything; the heuristic breathing_band_estimate is not a medical reading.
Determinism. Same input → same output. The DSP is hand-rolled (no rustfft/ndarray); ids come from a single IdGenerator; replay re-deserializes verbatim without re-validating.
#![forbid(unsafe_code)] in every crate except rvcsi-adapter-nexmon. rvcsi-node is deny(clippy::all). Keep files small, public APIs typed, input validated at boundaries.

Read the full file on GitHub · 38 lines

Changes

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 · 38 lines · 0 tokens per session scan A 2ee565880fdd

Subscribe to this mod's changes

rvcsi-engineer is an agent published in the GitHub repository ruvnet/rvcsi (27 stars, last pushed 14d ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 1,496 tokens. 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 2026-08-30.

Data and AI Languages Rust TypeScript channel-state-information csi edge-ai esp32 nexmon presence-detection

mcp-developer

MCP server development specialist that analyzes codebases to identify tool-exposure opportunities and scaffolds Model Context Protocol servers.

pjt222/agent-almanac · 27 tokens

logic-review

Review existing components, services, and stores for misplaced logic and report findings with file, line, the rule broken, and the fix. Use the logic-design agent to decide where something new should go.

AmadeusITGroup/otter · 43 tokens

tadpole-backend-specialist

Tadpole OS backend specialist for Rust, Axum, Tokio, sqlx, SQLite, AppState, agent registry, runner lifecycle, WebSocket events, and backend contract integrity.

DDS-Solutions/AI-TadPole-OS · 45 tokens

rust-cli-architect

Use this agent when you need to build complex Rust CLI applications, especially those involving ML/AI integration, system-level programming, or cross-platform development. Examples: Context: User wants to create a sophisticated command-line tool with multiple backends and safety features. user: 'I need to build a Rust…

wildcard/caro · 0 tokens

testing-engineer

Testing specialist for Rust, TypeScript, and Luau/Lua projects focused on comprehensive test coverage, test infrastructure, and quality assurance. Applies the project's language rules from the stack profile. Use PROACTIVELY when adding new functionality that requires tests, investigating test failures, or setting up…

bug-ops/claude-plugins · 137 tokens

critic

Adversarial critic for Rust, TypeScript, and Luau/Lua projects, specializing in finding logical gaps, flawed assumptions, scalability limits, and missing edge cases in architectural designs, implementation proposals, and ideas. Use PROACTIVELY after architecture design, before committing to an approach, or when a user…

bug-ops/claude-plugins · 131 tokens