youdotcom-oss/risk-analysis-server/extend
Claude Code and agent skill by youdotcom-oss/risk-analysis-server
- community
Checking that the SKILL.md file is still in the repository.
Clone this repo as a base and change its behavior - custom auth, a different model, extra tools, or a different report layout. Use when the built-in defaults (HMAC auth, qwen/qwen3.8-27b via OpenRouter, the three-section briefing) don't fit and you want the existing test scaffolding to keep you safe.
Install
npx vskill@latest install youdotcom-oss/risk-analysis-server/extendSKILL.md
Extend the risk-analysis server
Where things live
| What you want to change | File | |
|---|---|---|
| MCP tools / resources | src/mcp.ts | |
| HTTP auth, tenants, cron | src/server.ts | |
| Sweep orchestration (triage → deep dive) | src/pipeline/sweep.ts | |
| Query proposal / retrieval / synthesis | src/pipeline/deep-dive.ts | |
| Report Markdown assembly | src/pipeline/report.ts | |
| You.com access (search/contents tools) | src/services/you.ts | |
| Jev gates (triage, query validation, scoring) | src/services/jev.ts | |
| Sweep model selection (provider, model id) | src/model.ts | |
| Storage schema and task lifecycle | src/db.ts |
Steps
git clonethe repo andbun install.- Run
bun run check && bun testto confirm the baseline is green. - Change the file for your target behavior from the table above. Each module
has its own test in src/tests/ that pins its public behavior — change behavior by changing the test first, then the implementation.
- Verify with
bun run check && bun test. All green means your change is
behavior-complete against the pinned contracts.
Deploy your extension
The assets/ folder in this skill holds the container recipe this repo uses:
assets/Dockerfile— multi-stageoven/bun:1-alpineimage,/datavolume
for the SQLite database.
assets/docker-compose.yml— local orchestration with env wiring.
Copy them into your fork and set the env vars the compose file expects (RISK_JWT_SECRET, RISK_ALLOWED_HOSTS, RISK_CRON_SCHEDULE, plus the three API keys: YDC_API_KEY, TYPESAFE_API_KEY, OPENROUTER_API_KEY). Any host that runs a Bun container works.
Gotchas
- The stdio entry (
src/stdio.ts) is auth-free by design — do not add auth
there; the HTTP entry is src/server.ts.
- A failed cron handler with an unhandled rejection exits the process; keep
the per-profile try/catch in the cron handler when touching it.
buildMcpServerruns per request on the HTTP entry — do not capture
mutable state in the factory closure; tenant state flows through userId from the bearer token's sub.
Security
- Security scanPassed, 100 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingskills/extend/SKILL.md
Popularity
- 7-day trend score
- 0
Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/youdotcom-oss/risk-analysis-server/extend)