Skip to content

Open-source, community-driven agent harness

active 2026-05-232026-08-10 (UTC)

Partial coverage11,500 / 12,498 hourly files (92%) · 2 absent upstream · 994 failed, retryable2025-03-082026-08-10 (UTC)— sampled evenly across the window, so rankings and trends hold; absolute counts scale up.
Events
2.6K
Pushes
721
Pull requests
111
Issues
230
Stars
339
Forks
34

Activity over time

Daily event counts in the loaded window

Line chart, 80 days from 2026-05-23 to 2026-08-10. Pushes: 721 total, peak 56 in a day. Pull requests: 111 total, peak 26 in a day. Issues: 230 total, peak 35 in a day. Comments: 375 total, peak 66 in a day. Stars: 339 total, peak 59 in a day.

  • Pushes
  • Pull requests
  • Issues
  • Comments
  • Stars

Stars, PRs, issues and forks are under-captured in the later part of this window. GH Archive progressively stopped capturing non-push events during 2026 — −95% or worse by the end of the window. Every series here except Pushes fades for that reason, so a decline above reflects the archive, not this repository. Pushes stay reliable throughout, so read them, and the contributor counts derived from them, as the real signal. Data health has the measurements.

Top contributors

Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity

Recent activity

Latest issues, pull requests and releases

  • Issue comment#5295github-actions[bot]2026-08-08 13:11
    feat: add Mistral AI as a first-class provider route
  • Issue comment#4173Hmbown2026-08-08 02:47
    Architecture: de-hardcode model provider and tool registries
  • Issue comment#4394Hmbown2026-08-08 02:47
    Compaction: publish and enforce a structured survival contract
  • Issue comment#5244Hmbown2026-08-08 02:47
    Unknown model ids silently degrade to the 128K legacy context default — say so out loud
  • Issue#4754Hmbown2026-08-08 02:45
    Workflows follow-ups: /workflows live monitor, chat-side authoring, goal staging
  • Issue#4754Hmbown2026-08-08 02:45
    Workflows follow-ups: /workflows live monitor, chat-side authoring, goal staging
  • Issue#5037Hmbown2026-08-08 02:45
    Make web search and fetch a coherent, high-quality retrieval path
  • Issue#5041Hmbown2026-08-08 02:45
    Make notifications actionable, predictable, configurable, and consistently branded
  • Issue#5123Hmbown2026-08-08 02:45
    Agent spawn surface has too many knobs — labeled builder runs read-only and self-BLOCKED
  • Issue#5103Hmbown2026-08-08 02:45
    Rename DeepSeekClient / deepseek_client legacy internals to provider-neutral types
  • Issue#5093Hmbown2026-08-08 02:45
    Extract Responses dialect policy behind a language-neutral conformance harness
  • Issue comment#5249Hmbown2026-08-08 02:37
    Epic: v0.9.5 build-time lane — stop the monolith tax on every edit, commit, test, and release
  • Issue comment#5261Hmbown2026-08-08 02:37
    v0.9.5: extract the engine into `crates/core` + Thread/Session split
  • Issue comment#5146Hmbown2026-08-07 22:47
    Prompt teaches tool-call-is-proposal, then batch approval rejects the same call (owner session 1.9)
  • Issue#425Hmbown2026-08-07 22:46
    v0.9.3 Subagents: add resume_from continuation chains
  • Pull request#5281dependabot[bot]2026-08-07 21:16
  • Pull request#5278dependabot[bot]2026-08-07 21:15
  • Pull request#5275dependabot[bot]2026-08-07 21:14
  • Issue comment#5255aboimpinto2026-08-07 17:36
    Layer 5.3: Palette, completion, and discovery filtering
  • Issue comment#3310Hmbown2026-08-07 09:23
    v0.9.2: Split mcp.rs transports into focused modules
  • Issue#4978Hmbown2026-08-06 07:45
    频繁出现 Warn Anthropic API error (HTTP 400 Bad Request invalid_request_error): 'type' must be in ["enabled", "disabled", "auto"]
  • Issue comment#4376Hmbown2026-08-03 00:59
    v0.9.3: Add first-class llama.cpp (llama-server) provider support
  • Pull request#4874Hmbown2026-07-26 10:12
  • Issue comment#4855claude[bot]2026-07-26 05:13
    feat(fleet): add Oracle, a read-only high-reasoning advisor role (#4752)
  • Issue comment#4838Hmbown2026-07-26 04:13
    `codew model set` is a silent no-op for every non-DeepSeek provider (root `default_text_model` is DeepSeek-scoped)

Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 339 stars here means stars gained during the window, not the repo's star count.