AGENT SKILLS · 任务装备

值得安装的 Agent Skills

汇总公开仓库中已验证的 Skills。安装前先看清它能做什么、包内包含什么、热度如何，以及来源是否可靠。

目录范围: 已验证
Skills: 23,248
仓库: 3589
最近同步: 2026年11月4日

23,248 Skills

root-cause

Establish the minimal causal explanation for one reproducible or directly observed failure by building and falsifying competing mechanisms. Use when the missing outcome is diagnosis rather than report validity, review, or correction delivery.

1小时安装量: 1
GitHub Stars: 0
最近更新: 9月13日

se-triage

Assess a reported issue from supplied and bounded read-only evidence before implementation. Use to triage, validate, reproduce, classify, or choose the next step for a report. An explicit target authorizes relevant read-only investigation; provider comments and other mutations require separate authority.

1小时安装量: 1
GitHub Stars: 0
最近更新: 9月13日

seda-ticket

Break a supplied plan, specification, issue, conversation, or work description into consumable delivery tickets. Use when work needs small outcome-complete slices, explicit dependencies, acceptance criteria, and a clear startable frontier; exclude persistence, publication, execution, active review tracking, Git, and provider operations.

1小时安装量: 1
GitHub Stars: 0
最近更新: 9月13日

wo-pr

Babysit a GitHub PR or GitLab MR through CI, conflicts, and review feedback until ready for a human merge decision. Use for watching, following, or getting a PR ready, including an explicitly requested stack. Read-only requests stay read-only. Exclude independent review verdicts, approval, and merging.

1小时安装量: 1
GitHub Stars: 0
最近更新: 9月13日

yoruba-glossary

Decide additions or corrections to Yorùbá/English technical terminology and maintain an explicitly authorized glossary artifact. Use for actual term changes or glossary maintenance, not ordinary reuse of established terms.

1小时安装量: 1
GitHub Stars: 0
最近更新: 9月13日

coding-standards

Create, audit, or update a project's durable review standards. Use explicitly when accepted, project-specific coding rules require human judgment and cannot be enforced reliably by tooling.

1小时安装量: 1
GitHub Stars: 0
最近更新: 9月12日

deliver

Execute an accepted implementation specification sequentially in the current worktree through implementation, fresh quality repair, technical gates, and final product validation. Use explicitly to deliver already-settled software design; do not use to design the solution or manage parallel worktrees.

1小时安装量: 1
GitHub Stars: 0
最近更新: 9月12日

domain-language

Use the project's accepted domain vocabulary consistently across conversation, code, tests, documentation, design, and plans. Resolve and record only material vocabulary conflicts encountered during active work.

1小时安装量: 1
GitHub Stars: 0
最近更新: 9月12日

implementation-spec

Create an accepted, temporary implementation specification from settled software design before implementation begins. Use explicitly when the work needs a strictly ordered sequence of vertical slices and concrete validation evidence.

1小时安装量: 1
GitHub Stars: 0
最近更新: 9月12日

product-validation

Independently validate an accepted product journey through a real product interface and return Pass, Fail, or Inconclusive with direct evidence. Use after integrated implementation or whenever observable product behavior must be proven; do not use for code review, technical gates, or repair.

1小时安装量: 1
GitHub Stars: 0
最近更新: 9月12日

retro

Analyze a completed or paused agent run and preserve an evidence-backed retrospective for improving future work. User-invoked only; produce a temporary Markdown report without changing skills, project instructions, tooling, or code.

1小时安装量: 1
GitHub Stars: 0
最近更新: 9月12日

skill-guide

Explain available skills, compare their boundaries, or recommend what the user should invoke next. User-invoked only; provide guidance and example prompts without invoking, routing, or executing another skill.

1小时安装量: 1
GitHub Stars: 0
最近更新: 9月12日

add-model-price

Use when editing worker/src/constants/default-model-prices.json, packages/shared/src/server/llm/types.ts, pricing tiers, tokenizer IDs, or matchPattern regexes for OpenAI, Anthropic, Bedrock, Vertex, Azure, or Gemini model pricing.

1小时安装量: 1
GitHub Stars: 3.4万
最近更新: 9月10日

agent-setup-maintenance

Shared workflow for editing Langfuse's repo-owned agent setup under .agents/. Use when changing AGENTS files, shared skills, .agents/config.json, generated shim behavior, provider discovery paths, or install-time agent sync.

1小时安装量: 1
GitHub Stars: 3.4万
最近更新: 9月10日

analyze-cloud-costs

Analyze Langfuse Cloud infrastructure cost structure using Metabase cost marts. Use when asked about cloud spend, AWS versus ClickHouse cost splits, cost drivers by provider/service/usage type/account, daily cost per tracing event, infra cost dashboards, or cost regressions visible in Metabase.

1小时安装量: 1
GitHub Stars: 3.4万
最近更新: 9月10日

backend-dev-guidelines

Build or review Langfuse backend code. Use for tRPC routers, public REST APIs, BullMQ processors, services, middleware, Prisma or ClickHouse access, OpenTelemetry, Zod, environment configuration, or backend tests.

1小时安装量: 1
GitHub Stars: 3.4万
最近更新: 9月10日

changelog-writing

Shared workflow for writing Langfuse changelog entries after a feature is complete. Use when a branch is ready for merge and a changelog entry or changelog draft is needed.

1小时安装量: 1
GitHub Stars: 3.4万
最近更新: 9月10日

clickhouse-best-practices

MUST USE when reviewing ClickHouse schemas, queries, or configurations. Contains 28 rules that MUST be checked before providing recommendations. Always read relevant rule files and cite specific rules in responses.

1小时安装量: 1
GitHub Stars: 3.4万
最近更新: 9月10日

code-review

Review Langfuse code changes for correctness, regressions, and best practices.

1小时安装量: 1
GitHub Stars: 3.4万
最近更新: 9月10日

create-repo-agent

Design, implement, review, or harden Langfuse repo-owned autonomous agents. Use for LLM-powered GitHub Actions, scheduled or dispatched agents, agent-created PRs, prompts, allowlists, tokens, untrusted content, or self-updating instructions.

1小时安装量: 1
GitHub Stars: 3.4万
最近更新: 9月10日

datadog-query-recipes

Research Langfuse production telemetry with reusable Datadog queries. Use for tenant or project activity, API usage, queue behavior, spans, logs, metrics, or ad hoc measurements across production regions; pair with debug-issue-with-datadog for root-cause analysis.

1小时安装量: 1
GitHub Stars: 3.4万
最近更新: 9月10日

debug-issue-with-datadog

Establish root cause by combining Datadog telemetry with the Langfuse repo. Use when investigating or triaging a user report, Linear or GitHub issue, incident, or pasted production error.

1小时安装量: 1
GitHub Stars: 3.4万
最近更新: 9月10日

frontend-browser-review

Shared workflow for browser-based review of user-visible frontend changes in Langfuse. Use when a change affects UI behavior, layout, styling, navigation, or browser-visible regressions and should be checked with available browser automation before signoff.

1小时安装量: 1
GitHub Stars: 3.4万
最近更新: 9月10日

frontend-large-feature-architecture

Architect large or state-heavy Langfuse frontend features. Use for virtualized lists, large tables, controllers, Zustand stores, row selection, high-frequency state, rendering performance, or before adding useEffect, useMemo, useCallback, or form defaults derived from loaded data.

1小时安装量: 1
GitHub Stars: 3.4万
最近更新: 9月10日

NOASSERTION8 文件

查看详情 →