Agent skills for Devin Desktop

Use when deciding what lives outside a EuroSys paper's 12 technical pages — free reference pages, whether the current round permits…

Use when deciding whether a systems project belongs at EuroSys — the ACM SIGOPS-Europe flagship spanning OS, distributed systems, storage…

Evaluate and rank agent results by metric or LLM judge for an AgentHub session.

Bootstrap evaluators from production traces — by default propose online LLM-judge evaluators and, after you confirm, create them in…

Formal evaluation framework for Claude Code sessions implementing eval-driven development (EDD) principles.

Build the evaluation harness that gates every fine-tuning run — golden sets, per-failure-mode graders, judge calibration, and base-model…

Eval and improve a skill against golden cases — run the target skill blind in a fresh, context-free subagent on each example input, grade…

Use when testing Ralph's hat collection presets, validating preset configurations, or auditing the preset library for bugs and UX issues.

Guides evaluation of RAG pipeline retrieval and generation quality.

PluginEval quality methodology — dimensions, rubrics, statistical methods, and scoring formulas.

Look up EVE Online mission guides from eve-survival.org.

Hard-won ground rules for building EvenHub apps on Even Realities G2 smart glasses — SDK limits, audio/mic/VAD, TTS latency, rendering…

Use when documenting event objectives, requirements, and alignment across

Use when implementing decoupled communication between nodes — global EventBus autoload with typed signals

Event deduplication with canonical selection, reputation scoring, and hash-based grouping for multi-source data aggregation.

Message queue patterns with BullMQ, Kafka, RabbitMQ - saga, outbox, dead letter queue, exactly-once semantics.

Project directory path (default current directory)

Event Modeling facilitation technique — brainstorm events, identify commands and views, define aggregate boundaries, write Given-When-Then…

Complex event flow patterns for Valkyrie MoM scenarios.

Event prospecting skill. Takes a conference / event speakers URL, extracts the people, filters their companies against the user's ICP…

Store state as immutable event log instead of current values.