What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Session-trail is an open-source Claude Code plugin that turns selected work across coding-agent sessions into a timeline: prompts appear as ticks, milestones as nodes, and continuations between sessions as connecting curves. It is designed to complement a code graph, not replace one: a code graph shows project structure, while a session timeline can show how work unfolded and why certain decisions were made.
The project author introduced the idea in a September 17, 2026 DEV Community post. Its repository README documents the plugin, local data format, and command-line tools.
As an Amazon Associate I earn from qualifying purchases.
What the timeline shows
Session-trail gives each Claude Code session its own lane. Within a lane, time moves forward, while idle gaps are compressed so separate periods of activity can be viewed together. The timeline is organized around work and prompts rather than a map of files or dependencies.
- Nodes mark a decision or a completed piece of work.
- Curves connect sessions when later work continues an earlier line of work.
- Prompt ticks mark prompts recorded during a session.
Selecting a milestone opens its title and the recorded account of what happened, why, and how, along with nearby file touches, the session ID, and a path to the local transcript. Selecting a prompt tick shows the original prompt text.
#1 Best Overall
How prompts and milestones get recorded
The project describes two different capture paths. Hooks record sessions, prompts, file touches, and session titles. Milestones are selective: when the agent judges that a decision, reversal, completed unit, or cross-session continuation deserves one, it sends a brief to a Sonnet scribe subagent. The scribe drafts the milestone’s title and what, why, and how details; the rationale is meant to include rejected alternatives.
A session-end safety net can add an automatic node if work occurred without a milestone being recorded. This does not mean every important rationale is guaranteed to appear: milestone selection depends on the agent’s judgment, and the project documentation is not an independent accuracy or reliability evaluation.
Rank #2
The repository author summarizes the intended distinction this way: “The prompt ticks matter: they are the mechanical, complete record of what you asked, independent of the LLM’s judgment.” That is the project’s design claim, not an independently verified guarantee. Prompts are stored verbatim but truncated at 2,000 characters.
Where session-trail fits—and where it does not
A code graph and a session timeline answer different questions. A graph can help explain what the codebase contains and how its parts relate. Session-trail is intended to help reconstruct chronology: which prompts were made, which selected decisions or completed work were recorded, and how a later session continued an earlier one.
Rank #3
That makes the timeline useful as a project-history view, especially when a task crosses sessions. It is not a replacement for source control, code review, or a complete transcript archive. Its milestones are chosen summaries, while the prompt ticks and local transcript path offer ways to inspect more of the underlying interaction.
Install and open the viewer
The README lists Node.js 18 or newer as a requirement and says the project has no npm dependencies. In Claude Code, the documented setup and viewer command are:
Rank #4
- Run
/plugin marketplace add jsk4581/session-trail. - Run
/plugin install session-trail@session-trail. - Run
/session-trail:viewto open the timeline.
The repository also documents a manual note command and CLI commands for showing, searching, querying file history, serving the viewer, and checking the installation. Consult the README for exact command syntax and options.
Free tools Windows power users keep installed
One-click scans. No signup required.
Local history, sharing, and prompt privacy
Session-trail stores records in a project-local .session-trail/ directory. The README describes an append-only events.jsonl file and a derived graph cache. The history can be shared by committing that directory to the project, or kept personal by adding it to .gitignore.
Best Value
Because the event file can contain verbatim prompts, inspect it before committing. Prompts may include sensitive project context or other information you did not intend to share; truncation at 2,000 characters does not make them safe to publish.
The README says the viewer binds to 127.0.0.1 by default and that data endpoints require a per-start token. Serving on another interface is an explicit option, so check the repository’s instructions and your network exposure before enabling it.
What to consider before relying on it
- Chronology or structure: use the timeline to follow work over sessions; use a code graph when the question is how code elements relate.
- Exact interaction or selected rationale: prompt ticks record prompts, while milestone details depend on the agent deciding what merits a milestone and a subagent drafting it.
- Private record or team history: choose whether to ignore or commit the local history directory, and review prompt content before sharing.
- Documented behavior or measured outcome: the article and README describe intended features and implementation, but do not establish accuracy rates, time saved, adoption, or performance.
The central idea is not that a timeline can preserve every reason behind every change. It is that a chronological view of prompts and selected decisions may make the path through multi-session work easier to inspect alongside the code itself. What would you want a durable agent-session timeline to capture, and how do you currently keep track of work that spans many sessions?
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.




