Recipe
Health-check every connection in one pass
Every platform's health check run together: the morning coffee of agent operations.
IntentWhich connections are alive, degraded or dead, before work starts.
- 2
- Steps
- 6
- Tools
- 1
- Rules
- 1
- Failure modes
When to use it
Use it when
- Start of a session.
- After any OAuth or token event.
Do not use it when
- No specific contraindication beyond the preconditions below.
What it needs first
No single source is required: this one reads across whatever you have connected.
Preconditions
- The account is ticked in the dashboard, confirmed via list_accounts.
Inputs
window- Analysis window.Default: last 30 days.
Procedure
In order. Every tool named is one the gateway ships, and links to its reference.
Run each platform's health check.
Report per platform: healthy, degraded, needs human reconnection.
Decision rules
The observable condition, what it lets you conclude, and what takes the conclusion back.
When
A health check fails on auth
Conclude
Reconnection is a human step by design: hand back with the dashboard link, do not retry into a lockout.
Evidence every conclusion must carry
- Every figure carries its account, metric and window.
- Anything below a readable sample size is reported as unjudged, not as zero.
Where the agent stops
- The readout changes nothing. Any change it motivates goes through Safe Writes: preview, human confirmation, then apply.
How it goes wrong quietly
The cases where the analysis is wrong and still looks right. Read them before trusting a number.
- A passing health check does not guarantee scopes for every tool: it proves the connection, not the permissions.
What the answer contains
- 01Readout
- The figures, with windows and accounts named.
- 02Flags
- What deserves a deeper skill or a human decision.