shenli

Capture agent runs with checkpoints, effect receipts, and redacted trajectories.

Build independent, deterministic verifiers for agent-tool tasks.

Compare raw and official agent-tool surfaces using replicated AX tasks.

Design controlled coding-agent tasks for measuring developer-tool agent experience.

Turn replicated agent-experience findings into concise vendor reports.

Use when drafting, reviewing, restructuring, or updating software design documents, RFCs, engineering proposals, architecture proposals, implementation plans, or technical decision documents before or during a software project. Helps decide whether a design…

Use when finishing, pushing, opening, or preparing a pull request. Ensures relevant design, roadmap, architecture, operations, API, changelog, release, and implementation-status docs are updated according to repo instructions before the PR is considered…

Use when creating, auditing, reorganizing, renaming, consolidating, or updating documentation for infrastructure and developer products, including product thesis, design, architecture, roadmap, operations, API, SDK, runtime, reliability, and…

Use when designing a test plan for a distributed or stateful system — anything with persistence, replication, consensus, retries, idempotency, async messaging, multi-tenancy, or partial failure. Plans are claim-driven: investigate the product's claimed…

Use when running a previously designed distributed-systems test plan against a real or simulated cluster — driving fault injection, workload, chaos scenarios, linearizability / consistency runs, durability, partition, crash-recovery, upgrade, performance/SLO…