skills/core/aios-connect/SKILL.md
Wire a live connection to any tool. Applies api-vs-mcp-decision rules, researches the API, writes a reference doc, scaffolds a read-only wrapper, tests with a live call, and marks the bucket live in connections.md.
npx skillsauth add bereniketech/claude_kit aios-connectInstall 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.
Triggered with: /aios-connect <tool> or "connect me to <tool>".
If no tool given, read context/connections.md and recommend the highest-priority unwired bucket:
Tasks → Calendar → Comms → Revenue → Customer → Meetings → Knowledge.
Load .kit/skills/ai-platform/api-vs-mcp-decision/SKILL.md and apply its decision tree.
Key rules:
State the decision and reason in one sentence before proceeding.
Use WebSearch to find: auth method, base URL, rate limits, and the 5 most useful endpoints for AIOS use cases (read tasks, read events, read messages, read metrics, read files). Prefer official docs over community posts.
Create context/api-<tool>.md:
# <Tool> API Reference
## Auth
- Type: [OAuth2 / API key / service account / PAT]
- Env var: [VAR_NAME]
- Setup: [one-line steps to get the key]
## Rate Limits
- [requests/min or requests/day]
## Key Endpoints
### [Endpoint name]
GET https://api.example.com/v1/endpoint
Authorization: Bearer $VAR_NAME
Response: [what it returns in one line]
[repeat for up to 4 more endpoints]
## AIOS Use Cases
- Read today's tasks: [endpoint + key params]
- Check latest event: [endpoint + key params]
If the project has a lib/ directory, create lib/aios/<tool>.ts (adjust extension for project language):
export async function get<Tool>Data(params: Record<string, string> = {}) {
const url = new URL("https://api.example.com/v1/endpoint");
Object.entries(params).forEach(([k, v]) => url.searchParams.set(k, v));
const res = await fetch(url, {
headers: { Authorization: `Bearer ${process.env.TOOL_API_KEY}` },
});
if (!res.ok) throw new Error(`<Tool> API ${res.status}: ${await res.text()}`);
return res.json();
}
Rule: Wrappers are read-only by default. Write operations are added only when explicitly requested.
Make one read-only call. Report:
Connection test — <tool>
Endpoint: [what was called]
Status: [200 OK / error code]
Sample: [first field or item from response]
If the test fails, diagnose (wrong key, missing scope, wrong endpoint version) and fix before updating connections.md.
Update the relevant row in context/connections.md:
| Field | Value |
|---|---|
| Status | live if test passed · partial if test failed |
| Access method | lib/aios/<tool>.ts · or direct API · or MCP: <server-name> |
<Tool> connected.
Reference: context/api-<tool>.md
Wrapper: lib/aios/<tool>.ts
Status: live (test passed)
connections.md — [Bucket] bucket marked live.
Next: /aios-cadence to use this connection in a scheduled task.
testing
AUTHORIZED USE ONLY: This skill contains dual-use security techniques. Before proceeding with any bypass or analysis: > 1.
testing
Provide comprehensive techniques for attacking Microsoft Active Directory environments. Covers reconnaissance, credential harvesting, Kerberos attacks, lateral movement, privilege escalation, and domain dominance for red team operations and penetration testing.
development
Detects missing zeroization of sensitive data in source code and identifies zeroization removed by compiler optimizations, with assembly-level analysis, and control-flow verification. Use for auditing C/C++/Rust code handling secrets, keys, passwords, or other sensitive data.
development
Comprehensive guide to auditing web content against WCAG 2.2 guidelines with actionable remediation strategies.