PetalTrace
docs.petallabs.io
- Free plan
- Windows
- macOS
- Linux
- Web
- API
- Self-hosted
What PetalTrace is
PetalTrace is an open-source observability platform for developers working with AI agent workflows. It records prompts and completions, tool calls, token use, costs, and execution timelines. Full prompt capture can include system prompts, message history, tool definitions, and model responses. Developers can explore trace data in a React web interface or use the CLI, HTTP API, and MCP server. Search covers prompt and completion text, and real-time feeds show active runs. PetalTrace can compare two runs for differences in prompts, outputs, and costs, then replay a run with another model or temperature, or in mocked mode. It also accepts standard OpenTelemetry traces from instrumented applications, including those that do not use PetalFlow. PetalFlow integration adds graph structure, node inputs and outputs, and replay-capable snapshots. Runs, spans, and LLM interactions are stored in SQLite. The documented deployment options include building from source or running a release binary locally. PetalTrace is free, with an MIT-licensed public repository.
Who it suits
It suits developers who need to inspect and debug AI agent workflows, including teams using PetalFlow or applications instrumented with OpenTelemetry. Its local deployment and SQLite storage may also suit users who prefer to keep trace data in a local setup.
Good to know, and what to watch
- Captures prompts, tool calls, token use, costs, and timelines.
- Compares runs and supports replay.
- Accepts standard OpenTelemetry traces.
- Offers CLI, HTTP API, MCP server, and web UI.
- Authentication is future functionality and disabled by default.
- Documented trace storage uses SQLite.
Plans
Every plan and pricePetalTrace
Open source
Getting started
- Open the PetalTrace documentation website
- Build the software from Go source or download a release binary
- Run the PetalTrace daemon locally
- Use the React web interface, CLI, HTTP API, or MCP server
- Connect PetalFlow or send OpenTelemetry traces when applicable
Limits to check
The documented retention defaults keep runs for 30 days and failed runs for 90 days, with a maximum retention period of 365 days. Authentication is disabled by default and identified as future functionality.
The verdict
PetalTrace combines trace capture, comparison, and replay for AI agent workflows. Its disabled authentication is an important consideration for deployment planning.
Questions people ask
Is PetalTrace free?
Yes. The listed PetalTrace plan costs nothing, and the project is open source.
Which platforms does it support?
PetalTrace is listed for API, Linux, macOS, self-hosted, web, and Windows use.
What interfaces are available?
It provides a CLI, HTTP API, MCP server, and React-based web interface.
Can it replay an agent run?
Yes. Captured runs can be re-executed with different models or temperatures, including mocked, live, or hybrid modes.
Does it support OpenTelemetry?
Yes. It accepts standard OTLP traces from OpenTelemetry-instrumented applications, including applications that do not use PetalFlow.
How is data stored?
The documented trace store uses SQLite, with a default database path of ~/.petaltrace/data.db.
What the maker publishes
- Access methods
- The product exposes its capabilities through a CLI, HTTP API, and MCP server. source
- Authentication
- The configuration reference labels authentication as future functionality and shows it disabled by default. source
- Captured data
- It captures LLM prompts and completions, tool calls, token usage, costs, and execution timelines. source
- Capture modes
- PetalFlow integration offers minimal capture for latency, status, and token counts; standard adds prompts, completions, and tool I/O; full adds graph snapshots and edge data. source
- Capture modes
- PetalFlow integration offers minimal capture for latency, status, and token counts; standard adds prompts, completions, and tool I/O; full adds graph snapshots and edge data. source
- Cost tracking
- PetalTrace calculates costs for major providers and can track spending by workflow, provider, or model. source
- Debugging
- It can compare workflow runs for structural, content, and cost differences and replay runs in live, mocked, or hybrid modes. source
- Deployment
- The repository README documents building PetalTrace from source or downloading a release binary and running its daemon locally. source
- Installation
- The getting-started guide documents building PetalTrace from source with Go; the repository also links downloadable release binaries. source
- Integrations
- The MCP server lets AI agents query trace history, inspect prompts, analyze costs, compare runs, and trigger replays; the docs include Claude Code configuration. source
- Intended users
- The documentation describes PetalTrace as an observability platform for developers working with AI agent workflows. source
- Interfaces
- It provides a CLI, HTTP API, MCP server, and a React-based web UI for exploring traces, costs, and workflow graphs. source
- License
- The public GitHub repository identifies an MIT license. source
- Local storage
- The documented trace store uses SQLite, with a default database path of ~/.petaltrace/data.db. source
- Maker
- Petal Labs' GitHub organization describes the company as building modular, composable tools for agentic AI systems and lists its location as the United States of America. source
- MCP tools
- Its MCP server lets agents query traces, inspect prompts, analyze costs, compare runs, and trigger replays; the documentation shows Claude Code configuration. source
- OpenTelemetry
- PetalTrace accepts standard OTLP traces from any OpenTelemetry-instrumented application, including applications that do not use PetalFlow. source
- OpenTelemetry
- PetalTrace accepts standard OTLP traces over gRPC on port 4317 and HTTP on port 4318 from OTel-instrumented applications. source
- PetalFlow integration
- PetalFlow integration adds graph topology, node-level inputs and outputs, and replay-capable snapshots. source
- Product interface
- The repository README describes a React web UI for exploring traces, costs, and workflow graphs, alongside the CLI. source
- Prompt inspection
- Full prompt capture includes system prompts, message history, tool definitions, and LLM responses. source
- Purpose
- PetalTrace is an agent observability platform for inspecting AI agent workflows and their execution lifecycle. source
- Replay
- Captured runs can be re-executed with different models or temperatures, or in mocked mode. source
- Retention defaults
- Configuration defaults retain runs for 30 days, failed runs for 90 days, and allow a maximum retention period of 365 days. source
Checked against the maker's pages on 4 Oct 2026.