Set up the engineering workflow's per-repo context: an `## Agent skills` block in AGENTS.md/CLAUDE.md plus `docs/agents/` capturing the issue tracker (GitHub, GitLab, or local markdown), triage label vocabulary, and domain doc layout, with the full workflow skill map recorded as a reference inventory.

🔌

linear🔌Plugin

Use Linear issues, projects, and workflow state from engineering agents via the official Linear MCP server

🔌

web-bench🔌Plugin

WebBench benchmark runner - executes real-world browser tasks from the Halluminate/WebBench dataset, scores via LLM-as-judge, and produces evaluation reports

🔌

agent-web-interface🔌Plugin

Semantic browser interface for LLM agents - token-efficient page snapshots via Puppeteer/CDP, with a skill for live page interaction

🔌

regression-testing🔌Plugin

Regression testing scopes — define rerunnable confidence across changed and high-risk areas before automation selection

🔌

workflow-builder🔌Plugin

Scaffold complete Athena workflow packages with workflow definitions, plugin pins, marketplace registration, and validation gates

🔌

test-analysis🔌Plugin

Shared coverage planning, test-case generation, and spec review grounded in exploration evidence

🔌

playwright-automation🔌Plugin

Playwright execution layer for analysis, authoring, review, and flake repair; full workflows also require app-exploration and test-analysis