skills/golem-powers/_archive/context7/SKILL.md
Use when needing current API references, function signatures, or library usage patterns. Looks up documentation via Context7 API. Covers docs lookup, library documentation, API reference, how to use library. NOT for: web search (use WebSearch), project-specific code (read the codebase).
npx skillsauth add etanhey/golems context7Install this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Look up current library documentation using Context7 API.
Context7 provides up-to-date documentation and code examples for programming libraries. Prefer the Context7 MCP tools (mcp__context7__*) when available — they're simpler. Use this skill as a fallback when the MCP is unavailable or for scripted/batch lookups.
| Action | Script | Example |
|--------|--------|---------|
| Search for library | scripts/resolve-library.sh | ./resolve-library.sh react |
| Query documentation | scripts/query-docs.sh | ./query-docs.sh /vercel/next.js 'app router' |
| Show usage | scripts/default.sh | Auto-runs when skill loaded |
First, search for the library to get its Context7 ID:
./scripts/resolve-library.sh react
Output example:
## Library Search Results
| Name | ID | Snippets | Score |
|------|-----|----------|-------|
| React | /facebook/react | 1250 | 95 |
| React Native | /facebook/react-native | 890 | 92 |
Then use the library ID to query specific documentation:
./scripts/query-docs.sh /vercel/next.js 'how to use app router'
Output: Markdown-formatted documentation snippets with code examples.
The scripts automatically get the API key from:
CONTEXT7_API_KEY (if set)context7 in vault development, field API_KEYThe key is already stored in 1Password. No setup needed if you have op CLI authenticated.
If not using 1Password, set the environment variable:
export CONTEXT7_API_KEY="ctx7sk_your_key_here"
Get a key from context7.com → API settings.
If you want to use only this skill and reduce MCP overhead:
Edit your MCP config (~/.config/claude-code/mcp.json or mcp_config.json):
{
"mcpServers": {
"context7": {
"disabled": true
}
}
}
Or remove the entry entirely.
Edit ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) and remove or disable the context7 server.
This skill calls the Context7 REST API:
https://context7.com/apiGET /v2/libs/search?query=...&libraryName=...GET /v2/context?query=...&libraryId=...&type=txtWhen Context7 returns large documentation blocks (>100 lines), avoid reading them directly into Opus context. Instead use:
# Summarize large docs with external model
scripts/summarize-file.sh <file> "<what you need>" [gemini|cursor|kiro|haiku]
Or delegate to a subagent that writes a summary to a scratchpad file.
tools
The human-eval UX contract for Phoenix views: turn-by-turn scrollable replay (not a scorecard), hide-but-copyable IDs, collapsed thinking, identity chips, tool filters, tiny frozen starter datasets, mark-wrong-in-thread, mobile-first. Use when: building or reviewing ANY Phoenix/eval view, annotation UI, session replay, or human-grading surface. Triggers: phoenix view, eval UI, annotation view, session replay, human eval UX, grading interface. NOT for: Phoenix data pipelines/ingest (capture scripts have their own specs).
tools
macOS systems specialist — AppKit NSPanel architecture, launchd services, socket activation, MCP bridge resilience, syspolicyd, and high-frequency SwiftUI dashboards. Use when building menu-bar apps, LaunchAgents, debugging syspolicyd/Gatekeeper/TCC, resilient UDS/MCP bridges, or SwiftUI dashboards at 10Hz+.
development
Bulk LLM-judging protocol for fleet-dispatched verdict runs (KG cluster, eval harness). Use when: dispatching or running judge workers (J1/J2/RT), planning bulk-apply from verdict JSONL, or triaging evidence_degraded outputs. Triggers: judge fleet, bulk judge, R3 verdicts, kg-judge, RT gate, evidence_degraded. NOT for: single-item code review, Phoenix view UX (use phoenix-human-view), or non-judge eval pipelines.
development
Quiet-down protocol for sprint close: when the fleet wraps, delete ALL polling crons and monitors, send ONE final dashboard + ONE message, then go SILENT. Use when: fleet wraps, all workers done, overnight queue exhausted, sprint close, Etan asleep/away with nothing approved left. Triggers: fleet wrap, wrap the fleet, stand down, going quiet, sprint close. NOT for: mid-sprint monitoring (keep your loops), spawning a successor (use /session-handoff first).