ranvier-core

Hostile security and robustness audit of src/ assuming malicious traffic and 100x load. Use before releases, after large feature merges, or when the user asks for a security audit, adversarial review, or "what breaks under attack".

Holistic system audit for architecture drift, Hard Rules compliance, and async integrity across recently changed code. Use after completing a batch of tasks, before a release, or when the user asks for a health check or audit.

Semantic-correctness audit of hot-path code - enumerate the invariants a component must maintain, then hunt code paths that violate them (wrong answers, not crashes). Use after changing routing, eviction, accounting, or counter logic, before a release, or when metrics drift without any crash.

Orientation for a session starting work on this codebase - reading order, where things live, which skill to use for which task, and the non-negotiable constraints. Use at the start of a session, when unsure where to look, or when the user asks how to approach a task here.

Plan, run, and interpret Locust load benchmarks (mock-backend and real-GPU) including A/B routing experiments. Use when the user wants to measure cache-hit ratio, TTFT, or throughput, compare routing modes, chase a benchmark regression, or design a benchmark experiment.

Triage build or runtime failures. Use when the user pastes a compile error, link error, runtime crash, test failure, or error log, or mentions a build failure.

Post-implementation maintenance - write unit tests, sync documentation, and update BACKLOG.md. Use after code is implemented and reviewed, or when the user asks for tests or doc updates for a change.

Formalize a bug, audit finding, or incident root cause into a numbered Hard Rule in claude-context.md. Use after fixing a bug caused by a repeatable code pattern, after audits with findings rated 4+/10, or when the user says "make this a rule".

Start a structured implementation task with staged execution. Use when the user wants to implement a new feature, add functionality, or begin a coding task.

Structured incident response and root-cause analysis for production failures. Use when the user reports a production problem - crashes, latency spikes, OOM, hangs, wrong routing behavior - especially with logs, metrics, or benchmark evidence.

Scaffold a new service or component with correct Seastar lifecycle wiring - sharding shape, start/stop ordering, gates, metrics, config plumbing. Use when adding a new service, background worker, subscriber, or subsystem to the application.

Investigate performance bottlenecks with Seastar-aware analysis. Use when the user asks about performance, latency, throughput, optimization, or profiling.

Decompose a feature into atomic, verifiable implementation steps before coding. Use when the user describes a feature that spans multiple files or sessions, or asks for a roadmap, breakdown, or plan.

Refactor code with zero behavioral changes, preserving async semantics and shard affinity. Use when the user asks to restructure, extract, rename, move, or clean up code without changing what it does.

Run Hard Rules compliance review on changed code. Use when code has been written and needs review before commit, or when the user asks for a code review or self-review.

Brutally honest external-CTO assessment of project direction versus the stated goal. Use when the user asks where the project stands, whether it has drifted, what to cut, or for a state-of-the-project evaluation.

Pick the right verification ladder for a change - unit tests, sanitizers, validation suite, integration tests, fuzzing, lint - and emit exact Deferred Gate commands. Use before merging, when the user asks "how do I verify this", or when preparing the developer's Docker test checklist.