Fetching from the wire…
01
02
03
04
05
06
07
08
Public story · 2026-08-07 · high
The open-source Python tool breaks each task into tool calls, files, and tokens spent, all stored locally with no login.
Why now: agentacct was pushed August 7, built for scheduled multi-agent jobs. It turns a runaway token bill into something diagnosable after the fact, not a mystery on the invoice.
agentacct splits every AI agent task into a ledger of tool calls, files touched, tests run, and tokens spent, per the project's GitHub page.
That matters for anyone running scheduled agent jobs unattended, where a runaway loop can turn into a token bill with no explanation attached. agentacct's per-task breakdown turns that black box into something traceable back to a specific run.
The open-source Python tool, built by mikehasa, works across Claude Code, Codex, OpenCode and other agent runners.
It writes everything to a local-first dashboard instead of a hosted one. agentacct skips the login screen and sends no telemetry back to a server. Tool calls and file paths stay on the machine that ran the agent.
It's picked up 547 stars and ships under an MIT license. agentacct positions the no-telemetry design as its differentiator against hosted observability vendors.
Whether it stays telemetry-free as it adds features is the thing to watch.
Each link below shares sources, entities, or timing with this story.
312 stars and 24 forks within 24 hours of appearing August 11, MIT (GitHub). Notable for targeting three harnesses at once, which puts it in the emerging agent-agnostic UI layer category rather than single-vendor plugin territory. One day old, no releases, no independent verif...
QM went up under MIT license. Created July 29. As of the GitHub API check: 8,420 stars, 887 forks. Five days. YC uses it internally across accounting, legal, events, and engineering, including to build QM itself. Every employee and every Slack room gets its own scoped memory,...
A Chinese lab shipped a runtime that manages two American coding agents as subagents, and it went from repo creation to 145,439 stars in four days. deepseek-ai/deepseek-harness published dsh-v0.1.0-rc.7 at 12:01 UTC today, its first tagged release since the repo appeared on Au...
affaan-m/ECC (36.3k forks, MIT) bundles 67 agents, 284 skills, 94 legacy command shims, and "instincts", patterns learned from prior sessions with confidence scores that auto-recall when relevant, plus a .ecc/memory/ markdown vault that's explicitly cross-harness, so context s...
Spotify's Portal team published Xirp on August 10: a vendor-neutral agentic development environment that manages concurrent sessions across Claude Code, Gemini CLI, and Codex, each session isolated in its own git worktree so dozens of agents can work the same codebase without...
I check Product Hunt maybe once a week and usually regret it. Today's board is worth reading as market structure. The July 30 leaderboard: SKI at 277 upvotes (free voice input for Claude Code and Codex). AI Search Console at 249 (prompt analytics and citation mapping). Memmy A...
MindPattern daily
One email a day at 7 AM. Sources and a take on every story. Unsubscribe anytime.