Render deploy investigation brief
Read-only AI-assisted investigation brief for one Render deploy. It reads the service, deploy, recent deploys, service events, jobs, and logs through recued-core.render, then summarizes status, likely failure signals, nearby activity, and concrete next steps. It never triggers, cancels, rolls back, restarts, purges cache, creates jobs, reveals credentials, reads connection-info, mutates env vars/secrets, or calls arbitrary Render API routes.
How it works
Inspect the data fetches, transforms, gates, and output this recipe runs.
Process (24 steps)
normalized_service_id
trim
Trim whitespace from setting service id
normalized_deploy_id
trim
Trim whitespace from setting deploy id
normalized_owner_id
trim
Trim whitespace from setting owner id
service
?
deploy
?
deploys_raw
?
events_raw
?
jobs_raw
?
logs_raw
?
deploys
default
Apply default
events
default
Apply default
jobs
default
Apply default
logs
default
Apply default
event_count
count
Count items in events
job_count
count
Count items in jobs
log_count
count
Count items in logs
summarize
?
deploy_rows
slice
Take a subset of
event_rows
slice
Take a subset of
log_rows
slice
Take a subset of
deploy_table
to_table
Format results as a data table
event_table
to_table
Format results as a data table
log_table
to_table
Format results as a data table
card
to_summary
Format results as a summary card
Settings
Configurable at install. Defaults shown — change them anytime in Recued.
focus
setting
=
Summarize this Render deploy for a developer and on-call owner: current status, commit/image context, nearby deploys, service events, jobs, log signals, likely risks, and concrete next triage steps. Do not invent facts outside the Render data.
render
setting
=
owner id
setting
=
deploy id
setting
=
row limit
setting
=
20
max length
setting
=
900
service id
setting
=
Trust & control
What installing this recipe would let it do. Recued grants these permissions at install — review them there before approving.