Reliable AI agents.
Repeatable outcomes.
Agents on Jetty check their own work and improve until it’s right. Run a task once or a thousand times, get consistent results.
10 free runs · no credit card · no API key · then $20/mo for 10,000 runs
Trusted by teams at
Brickroad AI
For developers
Supercharge your agent with the Jetty skill
Create runbooks that check their own work, and monitor every run.
Start with a runbook
Each one is a real recurring job another operator runs every week. Fork it, tune it, schedule it.
Build repeatable tasks with Jetty.
Anatomy of a runbook
A runbook is a spec for your AI agent.
One portable Markdown file: your agent’s job, what “done” looks like, and how it checks its own work before finishing.
---
version: "1.0.0"
evaluation: rubric
agent: claude-code
model: claude-sonnet-4-6
snapshot: python312-uv
---
# ETL Pipeline Agent
## Objective
Fetch new events, enrich each record with
AI-generated summaries, and persist results
to the output table.
## REQUIRED OUTPUT FILES
| {{results_dir}}/validation_report.json |
| {{results_dir}}/summary.md |
| {{results_dir}}/enriched_events.csv |
## Parameters
| source_table | raw.events | Input |
| results_dir | /app/results | Output |
| batch_size | 100 | Records per batch |
## Step 1: Fetch Records
Query source_table for new rows since the
last checkpoint. Process in batches.
## Step 2: Enrich
For each batch, invoke the Summarizer skill
to generate a summary for each record.
## Step 3: Write Results
Persist enriched records to enriched_events.csv.
## Evaluation
| # | Criterion | 5 (Pass) | 1 (Fail) |
| 1 | Completeness | All rows enriched | Missing rows |
| 2 | Quality | Summaries coherent | Gibberish |
| 3 | Schema | Valid CSV output | Malformed |
Pass if score ≥ 4.0, no criterion below 3.
## Iteration
If evaluation fails, retry with bounded
iteration. Max 3 rounds.Two loops, one hill
The task fixes itself. The agent gets better.
Inner loop. Within a single run, the agent does the work, checks it against your standards, and refines until it passes. The run fixes itself.
Outer loop. Every run shows Jetty what worked and what didn’t. By week three, the agent knows your work better than your last assistant did.
Refinement isn’t a feature. It’s the engine.
Bring a job from your week.
By the end, you’ll have a repeatable task you can run a thousand times.
10 free runs · no credit card · no API key · then $20/mo for 10,000 runs
Backed by
