Comparison
sitrep vs AgentsView
Of every tool on these comparison pages, AgentsView is the closest neighbor: free, open-source, local-first, passive, cross-project - the same posture sitrep holds on every axis. The difference is what each one reads, and therefore what question it can answer. For "what did my agents actually do?", AgentsView is the better tool, plainly.
Disclosure: this page is written by the maker of sitrep. Read the framing with that in mind - and note that where the honest answer is "use AgentsView", this page says so.
The one-sentence difference
Your coding agents leave two trails behind. One is the session transcript - every prompt, tool call and token, logged as data. The other is the markdown they wrote on purpose - plans, analyses, decision logs, progress notes. AgentsView indexes the first trail and turns it into activity dashboards: what ran, when, at what cost. sitrep parses the second and turns it into a status view: where each project stands, what is stalled, what is waiting on you.
One is telemetry, the other is state. Telemetry tells you the agent worked for three hours on Tuesday; state tells you the migration is at phase 2 of 5 and blocked on a decision only you can make.
Side by side
| sitrep | AgentsView | |
|---|---|---|
| What it is | AI project status dashboard - a passive state layer over existing files | Session analytics for coding agents - dashboards over the transcripts they leave on disk |
| What it reads | The markdown agents wrote: plans, analyses, decision logs, progress notes | The logs agents left: session transcripts - prompts, tool calls, tokens - across many agent CLIs |
| Question answered | "Where do my projects stand, and what is waiting on me?" - forward-looking | "What did my agents do, when, and what did it cost?" - retrospective |
| Format demanded | None - reads what is already there | None - reads what is already there |
| Scope | Cross-project, across every repo and workspace | Cross-project, across every indexed agent and repo |
| State surfaced | Status, phases, open decisions, due dates, staleness against the repo | Activity, heatmaps, cost and token spend, session summaries and search |
| Writes to your files | Three guarded state operations in the document’s own convention; nothing else, ever | Nothing - read-only over the logs |
| Agent integration | MCP server exposing parsed state, read-only first | Reads the session files of many agent CLIs; the agents are the subject, not the client |
| License and price | Commercial desktop app, in development | Free, MIT-licensed open source |
| Local-first | Yes - no account, no network, no telemetry by default | Yes by default - indexes locally on your machine |
When to choose AgentsView
- You want to see and search what your agents did - sessions, timelines, heatmaps - across many different agent CLIs.
- Token and cost tracking matters: you want spend per project and per agent, measured from the logs.
- You need to find a past session - what was asked, what was tried - and replay the trail.
- You want free, MIT-licensed open source you can run today.
Session analytics is not on the roadmap - dashboards over transcripts are AgentsView's job. The most sitrep would ever take from the session trail is a timestamp: evidence that a project moved, feeding its staleness signals - never a surface built over the transcripts themselves.
When to choose sitrep
- The question is where things stand, not what happened: status, phases, open decisions, due dates - parsed from the docs themselves.
- Re-entry is the pain: opening a cold project after weeks and needing the state, not the history.
- You care whether the docs still tell the truth - staleness against repo activity is a signal you want surfaced, not something you reconstruct from timelines.
- Your state lives in markdown that no session produced - notes from other tools, hand-written decision logs, docs from before the agents arrived.
Using both
This is the most genuinely complementary pairing on these pages: same machine, same repos, two different trails read for two different questions. Run AgentsView to know what your agents did and what it cost; run sitrep to know where your projects stand and what needs you. Neither asks you to change anything about how you or your agents work, and both keep everythingon your machine.
sitrep is in development, launching first on macOS. Local-first - your files never leave your machine.