Skip to content

Agent-estimate

AI Agents & Assistants

A CLI tool that generates PERT‑style time estimates for AI‑agent tasks, showing expected duration, human‑equivalent cost, and model reliability limits.

Track releases GitHub
Python Latest v0.7.2 · 5h ago Security brief →

Features

  • Produces three‑point (optimistic/most‑likely/pessimistic) PERT timelines calibrated for AI agents
  • Compares estimated agent time to equivalent human effort (compression ratio)
  • Flags when an estimate exceeds a model's METR p80 reliability horizon
  • Supports multiple task types (coding, research, documentation, etc.) and automatic tier sizing
  • Planners wave‑parallel scheduling for fleets of agents

Recent releases

View all 11 releases →
Config change
v0.7.2 Bug fix

Documentation fixes + audit routing

Open
No immediate action
v0.7.1 Bug fix

Engine fixes + docs updates

Open
Review required
v0.7.0 Mixed
Auth

Frontend UI, App dev, Review mode, Audit logs

Open
No immediate action
v0.6.1 Bug fix

Corrected version reporting

Open
No immediate action
v0.6.0 Mixed

Model updates + README + examples

Open

Weekly OSS security release digest.

The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.

No spam, unsubscribe anytime.

About

Stars
3
Forks
1
Languages
Python Makefile
View on GitHub Documentation

Install & Platforms

Install via
pip

Similar tools

Claude-Autopilot
claude-code-from-scratch
Ait
AlysisAi/sylliptor
Aura

Beta — feedback welcome: [email protected]