Release safety
The platform reacts so you don’t get paged.
Mark any metric a guardrail with a direction and tolerance. When a ramp breaches it, Foredeck rolls the flag back to 0% on its own and tells you exactly which metric, effect size, and exposure counts justified the call.
ProIncluded from the Pro tier.
new-pricingauto-rolled-backA bad release, stopped in minutes — with the receipts.
What you get
Why Foredeck for guardrails & auto-rollback.
Guardrail metrics
Flag any metric as a guardrail with a direction and tolerance; it’s evaluated continuously against the live readout.
Automatic rollback
A breach rolls the flag to 0% as a first-class, auditable mutation — the same kill switch you’d hit by hand.
Rollback rationale
Every auto-rollback quotes the metric, effect size, confidence, and exposure counts — and can cite the exact trace or log line that regressed.
Ramp schedules
Step a rollout up on a schedule, with guardrails watching each step so a bad ramp never reaches 100%.
How it works
Three steps, one pipeline.
Mark a metric guardrail with a tolerance you’re willing to accept.
Foredeck watches the readout on every step of the ramp.
On a breach it rolls back and posts the rationale — for a human or an incident agent.
Better together
Works with the rest of the platform.
Experiments
FreeEvery rollout, automatically measured — experiments included free.
Explore Experiments →Tracing
FreeFollow one call across every service — spans stamped with the variation.
Explore Tracing →Logs
FreeSearch logs in release context — OTLP-native, correlated to flags.
Explore Logs →Agent Tools (MCP)
FreeManage Foredeck from Claude Code, Cursor, or your own agents.
Explore Agent Tools (MCP) →One object, one pipeline. The same rollout you ship is automatically measured, guarded, and — when it breaks — traceable, because every signal is joined to the flag and variation the unit was in.
Get in