Lunar Command
Official@LunarCommand · United States of America
Agentic Narrative Company
Agent Skills by Lunar Command
Showing 6 vetted skills indexed across 1 GitHub repositories.
LunarCommandlangfuse
Query Langfuse traces, observations, sessions, scores, and prompts across legacy and v4 APIs.
LunarCommandhyperdx
Query HyperDX logs and traces with Lucene syntax via a bundled CLI script.
LunarCommandmutation-test
Validates test assertions by mutating source code and confirming the test suite fails.
LunarCommandadversarial-review
Generates verified code and spec review findings using independent multi-lens adversarial agents.
LunarCommandfeature-planning
Generates requirements analysis and implementation plans with two human approval gates before coding.
LunarCommandpr-review
Triages GitHub PR review threads one at a time with approval-gated replies and resolution.
Frequently Asked Questions About Lunar Command
FAQPage SchemaWhat tasks can I accomplish using Lunar Command's skills?▼
You can query Langfuse traces, prompts, and scores; search HyperDX logs and traces with Lucene syntax; verify tests genuinely assert behavior via mutation testing; run adversarial multi-lens code reviews; plan features from requirements documents; and address pull request review comments.
Who are Lunar Command's skills designed for?▼
Software engineers, SREs, and QA-focused developers who debug incidents, review pull requests, and validate test suites. The skills target practitioners working with Langfuse observability data, HyperDX log investigation, PR review threads, and requirements-driven feature implementation.
How do the observability skills work at runtime?▼
The HyperDX skill queries logs exclusively through the bundled hdx_query.sh script using Lucene syntax—never via curl or direct API calls. The Langfuse skill works against self-hosted and cloud deployments, supporting both legacy v1 REST (server major <= 3) and v2/v3 read APIs (server major >= 4).
How does the mutation-test skill verify test quality?▼
It breaks the behavior a test claims to cover and confirms the test goes red, providing evidence rather than trusting a green run. It can scope verification to a PR or diff by shortlisting changed lines in a throwaway checkout, without mutating your actual codebase.
What makes the adversarial-review skill different from standard code review?▼
It generates findings rather than triaging them, running specialized lenses that look for what breaks instead of what code does. Each finding is verified by refutation before surfacing, ranked by severity, and PR-scale reviews escalate to independent multi-agent lens execution.