Sediment vs Entire
Use Entire to review coding-agent sessions and the commits they produced. Sediment preserves coding-agent evidence on your infrastructure for reports, context agents can query, and proprietary training datasets.
Updated September 28, 2026 · Competitor sources reviewed September 2, 2026
Main differences
| Dimension | Entire | Sediment |
|---|---|---|
| What you record | Agent transcripts linked to commits as checkpoints. Checkpoint docs | Model calls, developer decisions, file edits, commits, and CI results. Sediment docs |
| What you get | Read checkpoints on your machine. Search pushed, indexed records through its hosted service after login. Inspection docs Search docs | Reports on agent work, captured context that agents can query, and training datasets with their source evidence. Agent work reports Evidence retrieval Training exports |
| Where you run it | Local CLI and Git storage. Keep checkpoints on your machine or push them to a remote. Checkpoint storage | Capture, storage, evidence reads, reports, and exports run on your infrastructure. Your agent’s model endpoint controls inference traffic. Network boundaries |
Claude Code is stable; the other built-in integrations are in preview. Token reporting, subagent capture, and session resume vary by agent. Agent support
Using both
Sediment returns exact captured messages, tool calls, and tool results from a previous session. A pi agent can query them itself, with a token limited to the sessions you grant, or an operator can prepare an evidence packet with the CLI. No model selects the evidence. Evidence retrieval
They use separate Git storage locations: Entire checkpoints and Sediment commit notes. Sediment keeps its full records outside Git. Checkpoint storage Commit links Neither vendor documents an integration, so test the combination in your workflow.