Skip to content
botshub

agentpace

Coding agents

Visit agentpace

agentpace estimates P50, P80 and P90 completion times for plans of coding-agent tasks using measured runs, while reporting the human part separately. It also provides a view of completed run durations and breakdowns by model, task kind and risk.

About agentpace

agentpace is a command-line tool that estimates how long plans of coding-agent tasks will take, using measured agent runs instead of human-team estimates. It returns P50, P80 and P90 completion times and reports the human-equivalent time separately.

The details here come from the vendor's own site, and I found no independent coverage. The page names no organization, and its GitHub links point to mohamedzhioua/agentpace, so that appears to be the author. It says it ships a dataset of 3,383 completed runs, though another passage cites 3,430, so the count is inconsistent. It logs runs locally from Claude Code subagents and Codex and says it makes no external requests. The page states no license. It installs with uv or as agent skills through npx.

Key features

  • P50, P80 and P90 estimates
  • Task dependencies and agent cap
  • Built-in run dataset
  • Local run logging
  • Separate human-time line

Frequently asked questions about agentpace

What is agentpace?
agentpace estimates P50, P80 and P90 completion times for plans of coding-agent tasks using measured runs, while reporting the human part separately. It also provides a view of completed run durations and breakdowns by model, task kind and risk. It is listed on botshub under coding agents.
What can agentpace do?
Key features include p50, P80 and P90 estimates, task dependencies and agent cap, built-in run dataset, local run logging and separate human-time line.
What are alternatives to agentpace?
Other coding agents on botshub include AgentPlaza, Capy, OASIS and Poolside. See all alternatives to agentpace.
Report this listing
Is this your product? Claim this listing