Read-only Railway deployment investigation brief. It reads one deployment plus bounded build and runtime logs through recued-core.railway, then asks the AI summarizer for likely failure signals, release state, and concrete next checks. It does not restart, stop, cancel, redeploy, roll back, mutate variables, delete resources, or call arbitrary Railway GraphQL.

How it works 15 steps

Inspect the data fetches, transforms, gates, and output this recipe runs.

Process (15 steps)

■ target_deployment_id trim

Trim whitespace from setting deployment id

■ normalized_log_filter trim

Trim whitespace from setting log filter

■ deployment ?

■ runtime_logs_raw ?

■ build_logs_raw ?

■ runtime_logs default

Apply default

■ build_logs default

Apply default

■ runtime_log_count count

Count items in runtime logs

■ build_log_count count

Count items in build logs

■ summarize ?

■ runtime_rows slice

Take a subset of

■ build_rows slice

Take a subset of

■ runtime_table to_table

Format results as a data table

■ build_table to_table

Format results as a data table

■ card to_summary

Format results as a summary card

Settings 7 configurable

Configurable at install. Defaults shown — change them anytime in Recued.

focus setting = Summarize the Railway deployment status and logs. Identify likely failure signals, rollout state, rollback/redeploy availability, and concrete next checks. Quote only short log fragments and do not invent facts outside the Railway data.

railway setting =

log limit setting = 100

row limit setting = 25

log filter setting =

max length setting = 1000

deployment id setting =