Changelog

What's new

Selected milestones from the ClevAgent agent workspace build.

v0.13

Export your security records as one Excel workbook. The new Audit page shows what is in it.

v0.12

Your agents can now work as a team. Set up an orchestration once and ClevAgent passes each finished answer to the next agent, so you are not copying and pasting between them.

v0.11

New Security Rules. ClevAgent warns on mishandled secrets, blocks prompt injection, stops cleanups that would erase unsaved work, and pauses the agent before it changes its own settings, opens hidden access, or runs a destructive command. Security Mode goes further and blocks secrets before they leave your machine.

v0.10

v0.09

Four more optimization Rules. ClevAgent now flags crowded document folders, code changed without a check that passed, the agent retrying the same failing fix, and very large files pulled into context whole.

v0.08

New Max plan: run up to 5 agents with the same ClevAgent guidance for $49/month.

v0.07

The Mac app now runs natively on both Apple Silicon and Intel Macs. Hit Download and pick the version that matches your Mac.

v0.06

Character Shell preview added to the agent workspace. Agents can now appear as visual characters with speech bubbles, action states, and tool-aware status updates while they work.
ClevAgent guidance now appears as friendly intervention cards inside the workspace, keeping important nudges visible without breaking the flow.
Task complete and error states now have dedicated visual cards, making agent results and issues easier to follow at a glance.

v0.05

Gemini CLI added to the onboarding wizard. Pick Claude Code, Codex, or Gemini at first run, and ClevAgent supervises whichever you choose.

v0.04

Pre-launch evidence updated from the current working-session replay: $1,308 estimated savings per 100 real sessions, 44,882 LLM calls analyzed, and 2,441 waste signals caught.
Our Rules expanded with rule-by-rule detection and savings methodology for duplicate reads, stale readbacks, file organization, long files, and memory hygiene.
Agent workspace preview refreshed with current app screenshots showing live ClevAgent guidance and optimization history.

Changelog

Selected milestones from the ClevAgent agent workspace build.

v0.13

Export your security records as one Excel workbook. The new Audit page shows what is in it.

v0.12

Your agents can now work as a team. Set up an orchestration once and ClevAgent passes each finished answer to the next agent, so you are not copying and pasting between them.

v0.11

New Security Rules. ClevAgent warns on mishandled secrets, blocks prompt injection, stops cleanups that would erase unsaved work, and pauses the agent before it changes its own settings, opens hidden access, or runs a destructive command. Security Mode goes further and blocks secrets before they leave your machine.

v0.10

v0.09

Four more optimization Rules. ClevAgent now flags crowded document folders, code changed without a check that passed, the agent retrying the same failing fix, and very large files pulled into context whole.

v0.08

New Max plan: run up to 5 agents with the same ClevAgent guidance for $49/month.

v0.07

The Mac app now runs natively on both Apple Silicon and Intel Macs. Hit Download and pick the version that matches your Mac.

v0.06

Character Shell preview added to the agent workspace. Agents can now appear as visual characters with speech bubbles, action states, and tool-aware status updates while they work.
ClevAgent guidance now appears as friendly intervention cards inside the workspace, keeping important nudges visible without breaking the flow.
Task complete and error states now have dedicated visual cards, making agent results and issues easier to follow at a glance.

v0.05

Gemini CLI added to the onboarding wizard. Pick Claude Code, Codex, or Gemini at first run, and ClevAgent supervises whichever you choose.

v0.04

Pre-launch evidence updated from the current working-session replay: $1,308 estimated savings per 100 real sessions, 44,882 LLM calls analyzed, and 2,441 waste signals caught.
Our Rules expanded with rule-by-rule detection and savings methodology for duplicate reads, stale readbacks, file organization, long files, and memory hygiene.
Agent workspace preview refreshed with current app screenshots showing live ClevAgent guidance and optimization history.