---
title: "agents-runtime"
description: "Runtime analysis skills: performance, startup, concurrency, production parity."
canonical: https://agentpluginsdirectory.com/plugins/agents-runtime
last-updated: 2026-10-06
---

# agents-runtime
Runtime analysis skills: performance, startup, concurrency, production parity.
- Slug: agents-runtime
- Publisher: skillshop-ostyles
- Repository: https://github.com/skillshop-ostyles/skillshop-agents
- Manifest: plugins/agents-runtime/plugin.json
- Version: 1.0.0
- License: MIT
- Category (editorial): code-review
- Skills: 14 (cache-effectiveness-auditor, concurrency-hazard-scanner, dead-code-at-runtime, dependency-runtime-availability, error-propagation-mapper, mock-production-gap, process-lifetime-tracker, prod-mirror, repro-builder, runtime-type-mismatch, schema-query-mismatch, shutdown-gracefulness, side-effect-ordering, startup-profile-analyzer)
- MCP servers: 0
- Stars: 0
- Repository created: 2026-07-22
- Repository last pushed: 2026-08-22
- Publisher type: User
- Listing: https://agentpluginsdirectory.com/plugins/agents-runtime
- Schema: https://agent-plugins.org/schemas/1.0.0/plugin.schema.json

## Skills

- cache-effectiveness-auditor: Cache effectiveness auditor: inventory every caching pattern, extract strategy (TTL/invalidation/key design), LLM judges if each cache is effective or harmful. Read-only. Trigger: /cache-audit
- concurrency-hazard-scanner: Concurrency hazard scanner: map shared mutable state across async boundaries, LLM judges each pattern as safe/racy/deadlock-prone. Read-only. Trigger: /concurrency
- dead-code-at-runtime: Dead code at runtime detector: find feature flags, date gates, and env checks that make code unreachable in practice. Read-only. Trigger: /dead-runtime
- dependency-runtime-availability: Dependency runtime availability checker: find dynamic imports and runtime resource references that fail in production. Read-only. Trigger: /runtime-deps
- error-propagation-mapper: Error propagation mapper: trace every error from origin through handling blocks to surface, LLM classifies each path as monitored/silent/dangerous. Read-only. Trigger: /error-map
- mock-production-gap: Mock-production gap detector: compare test mocks against real implementations, LLM judges dangerous divergences. Read-only. Trigger: /mock-gap
- process-lifetime-tracker: Process lifetime tracker: map every process/service/daemon, trace shutdown paths, LLM judges graceful shutdown readiness. Read-only. Trigger: /lifetime
- prod-mirror: Production behavior mirror: ingests exported log files (text or JSON lines), statistically condenses them (frequencies, error rates, hot paths), extracts the code's expectations (log statements, catch blocks, routes), then has the LLM report the deltas, dead features, swallowed errors firing daily…
- repro-builder: Turns a vague bug report into a minimal, runnable reproduction: extracts hypotheses from the report text, snapshots the environment, generates a repro test/script, EXECUTES it and iterates (max 5 attempts) until the bug demonstrably reproduces, or documents precisely which information is missing.…
- runtime-type-mismatch: Runtime type mismatch detector: find every runtime type assumption, LLM judges which will fail in production. Read-only. Trigger: /type-mismatch
- schema-query-mismatch: Schema-query mismatch detector: compare every query pattern against the declared DB schema, LLM judges production risk. Read-only. Trigger: /schema-query
- shutdown-gracefulness: Shutdown gracefulness analyzer: check if shutdown hooks actually drain, flush, and complete in-flight work. Read-only. Trigger: /shutdown
- side-effect-ordering
- startup-profile-analyzer

Descriptions come from the frontmatter of each SKILL.md, punctuation lightly normalized.