Skip to content

Roboco

v0.2.0 Feature

This release adds 3 notable features for engineering teams evaluating rollout.

Published 1d AI Agents & Assistants
✓ No known CVEs patched
Read the diff → Tool health → What is this tool? →

✓ No known CVEs patched in this version

Topics

agentic-ai agentic-workflow claude state-machine

Summary

AI summary

Agents can now run the full quality gate, fixing lint/type/complexity issues from merging unchecked.

Changes in this release

Feature Medium

RoboCo queues work instead of dropping it when a provider returns 429.

RoboCo queues work instead of dropping it when a provider returns 429.

Source: llm_adapter@2026-06-12

Confidence: high

 —
Feature Medium

Agents park themselves via `i_am_blocked(reason="rate_limited")` during provider rate limits.

Agents park themselves via `i_am_blocked(reason="rate_limited")` during provider rate limits.

Source: llm_adapter@2026-06-12

Confidence: high

 —
Feature Medium

Background probe-and-resume loop revives parked agents when rate limits lift.

Background probe-and-resume loop revives parked agents when rate limits lift.

Source: llm_adapter@2026-06-12

Confidence: high

 —
Feature Medium

Control panel displays a live rate‑limit banner showing current limits.

Control panel displays a live rate‑limit banner showing current limits.

Source: llm_adapter@2026-06-12

Confidence: high

 —
Feature Medium

New `/ws/system` WebSocket stream forwards system events to the panel in real time.

New `/ws/system` WebSocket stream forwards system events to the panel in real time.

Source: llm_adapter@2026-06-12

Confidence: high

 —
Feature Medium

Token usage per agent session is captured and persisted for daily rollups.

Token usage per agent session is captured and persisted for daily rollups.

Source: llm_adapter@2026-06-12

Confidence: high

 —
Feature Medium

Dashboard shows real per‑agent token costs with provider‑aware pricing (Anthropic priced correctly, local/Ollama runs $0).

Dashboard shows real per‑agent token costs with provider‑aware pricing (Anthropic priced correctly, local/Ollama runs $0).

Source: llm_adapter@2026-06-12

Confidence: low

 —
Bugfix Medium

Token‑usage capture now correctly populates per‑session counters, fixing zero values in the dashboard.

Token‑usage capture now correctly populates per‑session counters, fixing zero values in the dashboard.

Source: llm_adapter@2026-06-12

Confidence: high

 —
Bugfix Medium

Control‑panel rate‑limit endpoint shape and duplicated `/ws/ws/system` WebSocket path are corrected.

Control‑panel rate‑limit endpoint shape and duplicated `/ws/ws/system` WebSocket path are corrected.

Source: llm_adapter@2026-06-12

Confidence: high

 —
Bugfix Medium

Agents can now run the full quality gate with `dev` extra installed, preventing unchecked lint/type/complexity issues.

Agents can now run the full quality gate with `dev` extra installed, preventing unchecked lint/type/complexity issues.

Source: llm_adapter@2026-06-12

Confidence: low

 —
Bugfix Medium

CEO reject conduit and main PM routing issues are resolved.

CEO reject conduit and main PM routing issues are resolved.

Source: granite4.1:30b@2026-06-12-audit

Confidence: low

 —
Bugfix Low

Provider‑aware model pricing is fixed: Opus priced at $5/$25 per 1M tokens, non‑Anthropic models no longer warn or misprice.

Provider‑aware model pricing is fixed: Opus priced at $5/$25 per 1M tokens, non‑Anthropic models no longer warn or misprice.

Source: granite4.1:30b@2026-06-12-audit

Confidence: low

 —
Bugfix Low

Control‑panel logo and `/public` assets now serve correctly without 500 errors after fixing file ownership for the non‑root runtime user.

Control‑panel logo and `/public` assets now serve correctly without 500 errors after fixing file ownership for the non‑root runtime user.

Source: granite4.1:30b@2026-06-12-audit

Confidence: low

 —
Bugfix Low

Agent idle deadlock and lifecycle hardening improvements prevent deadlocks during agent inactivity.

Agent idle deadlock and lifecycle hardening improvements prevent deadlocks during agent inactivity.

Source: granite4.1:30b@2026-06-12-audit

Confidence: low

 —
Full changelog

This release makes RoboCo resilient to LLM provider rate limits, gives the dashboard real per-agent token costs, and pushes system state to the panel in real time — plus the fix that finally lets the agents gate their own work.

Highlights

Provider rate-limit handling. When a provider returns a 429, RoboCo no longer drops work or spins. A Redis-backed tracker flags the provider, the spawn gate queues (never drops) further work for it, and affected agents park themselves via i_am_blocked(reason="rate_limited"). A background probe-and-resume loop clears the limit and revives the parked agents the moment it lifts — escalating to the CEO only if probes keep failing. The control panel shows it all live via a rate-limit banner.

Token usage & cost analytics. Every agent session's token usage is now captured from the Claude Code transcript and persisted into per-session rows and daily rollups, with provider-aware pricing (Anthropic models priced correctly; local/Ollama runs intentionally $0). The usage dashboard finally shows real numbers instead of zeros.

/ws/system operator stream. A new system-wide WebSocket stream and event bridge forward system events (starting with the rate-limit lifecycle) from the event bus to the panel in real time — the reusable foundation for future live operator views.

Fixed

  • Agents can now run the full quality gate. Workspaces install the project's dev extra, so a spawned agent's environment carries ruff/mypy/xenon/pytest —
    closing the gap that previously let lint/type/complexity issues merge unchecked.
  • Token-usage capture (the dashboard recorded zeros because nothing populated the per-session counters).
  • The panel rate-limit endpoint shape and the doubled /ws/ws/system WebSocket path.
  • Control-panel logo and all /public assets returning 500 (the image copied them without chowning to the non-root runtime user).
  • Provider-aware model pricing (Opus corrected to $5/$25 per 1M tokens; non-Anthropic models no longer warn or mis-price).

What's Changed

  • Fix: CEO reject conduit and main pm routing by @rennf93 in https://github.com/rennf93/roboco/pull/91
  • Fix: agent idle deadlock and lifecycle hardening by @rennf93 in https://github.com/rennf93/roboco/pull/96
  • [499f9eb1] Token Usage & Cost Analytics — Full-Stack Instrumentation, Persistence, and Visualization by @rennf93 in https://github.com/rennf93/roboco/pull/90
  • Chore: reduce analytics complexity by @rennf93 in https://github.com/rennf93/roboco/pull/100
  • [aaac85d2] Rate limit guardrails for Anthropic and Ollama providers by @rennf93 in https://github.com/rennf93/roboco/pull/104
  • Fix: rate limit real probe by @rennf93 in https://github.com/rennf93/roboco/pull/110
  • [4865ff8b] Add WebSocket support to the usage dashboard by @rennf93 in https://github.com/rennf93/roboco/pull/115

Full Changelog: https://github.com/rennf93/roboco/compare/v0.1.0...v0.2.0

View diff on GitHub

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.

Share this release

Share on X Share on Bluesky

Track Roboco

Get notified when new releases ship.

Sign up free

About Roboco

All releases →

Related context

Beta — feedback welcome: [email protected]