← Back to all sparks
D

Dosu

AI-ASSISTANTS
Velocity6.3

AI knowledge agent that answers questions and triages issues from your codebase

Dosu moved from maintaining your repo to measuring what your coding agents actually did.

agent-observabilitycost-trackingcoding-agentsdocumentationlocal-firstdevtools
Current state
Dosu started as an AI teammate for repository upkeep — documentation freshness scoring, stale-issue triage, templated release notes — and spent the spring making that configurable through Libraries and Agents. It dropped its waitlist in July and added usage analytics so teams could see its impact. Decant is a departure: a local tool that reads Claude Code and Codex session logs and reports what those agents did and what they cost.
Where it's heading
The through-line is that Dosu keeps productizing the parts of agent work that are hard to see — first whether docs were stale, then whether Dosu itself was earning its place, now whether anyone's coding agents are. Building Decant to run locally rather than as a hosted service sidesteps the objection that session logs are sensitive, which suggests it is aimed at teams that would not upload them. The feed is excerpt-only, so the depth of the tool is not visible from the changelog alone.
Prediction
The obvious next step is connecting Decant's per-session cost data back to Dosu's own analytics, so a team can compare what its coding agents spend against the maintenance work Dosu absorbs — though the entries do not yet confirm that direction.

Recent moves

  1. 1d ago

    Introducing Decant: Insights for your Claude Code and Codex sessions

    ⚡ SPARK

    Decant is Dosu's first product aimed at the coding agents themselves rather than the repository they work in, turning Claude Code and Codex session logs into usage and cost numbers. It extends the measurement instinct behind Dosu's docs-freshness scores and usage analytics into a category the company had not previously entered.

    View source ↗
  2. 22d ago

    July Dosu Drop: Addition by Subtraction

    Dosu removed its waitlist, simplified the Knowledge Cache tools, and added impact stats. Opening access is the notable half — it marks the point where Dosu stopped rationing onboarding, which is a prerequisite for launching adjacent products like Decant.

    View source ↗
  3. 1mo ago

    June Drop: Introducing Libraries and Agents

    An overhaul of how Dosu is configured, reorganized around Libraries and Agents. This is the structural change that lets the product carry several distinct behaviors instead of one fixed assistant.

    View source ↗
  4. 1mo ago

    Automate recurring work with Dosu Templates

    Templates turn recurring judgment-heavy work — release notes, status updates, triage — into reusable definitions Dosu drafts from connected sources. It is the same pattern the company keeps applying: take a repeated manual chore and make it a configurable artifact.

    View source ↗
  5. 2mo ago

    A stale AGENTS.md is worse than no AGENTS.md

    A blog post on keeping AGENTS.md files current, citing Cloudflare's internal platform. Positioning content for the docs-freshness product rather than a shipped change.

    View source ↗
  6. 2mo ago

    May Drop: New usage analytics to see Dosu's impact

    Usage analytics arrive so teams can quantify Dosu's impact, alongside bulk doc generation and agent-driven setup. The analytics thread started here and runs directly to Decant three months later.

    View source ↗