Skip to content

Token Efficiency For AI Agents

active 2026-05-072026-08-14 (UTC)

Complete coverage26,424 / 26,424 hourly files (100%) · 2 absent upstream2023-08-152026-08-19 (UTC)
Events
482
Pushes
221
Pull requests
13
Issues
19
Stars
3
Forks
0

Activity over time

Daily event counts in the loaded window

Line chart, 100 days from 2026-05-07 to 2026-08-14. Pushes: 221 total, peak 19 in a day. Pull requests: 13 total, peak 4 in a day. Issues: 19 total, peak 3 in a day. Comments: 10 total, peak 2 in a day. Stars: 3 total, peak 1 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

ContributorContributionsPushesPRsComments
anilmurty17515062
anshss756851
greptile-apps[bot]9007
github-actions[bot]4200
sjhddh1010
kelter-antunes1010
auspexlabs1100

Recent activity

Latest issues, pull requests and releases

  • Pull request#708anshss2026-08-07 18:29
  • Pull request#707anshss2026-08-07 17:39
  • Issue comment#707greptile-apps[bot]2026-08-07 17:14
    Remove write-budget gate: apply is offered whenever a finding is apply_capable
  • Pull request#707anshss2026-08-07 17:12
  • Pull request#703anshss2026-08-07 17:04
  • Issue#673anilmurty2026-07-31 20:00
    tj context output is repetitive — dedup the percentages and the per-file 'Reference…' instruction
  • Pull request#636anshss2026-07-28 10:59
  • Issue#615github-actions[bot]2026-07-27 19:12
    test_ping fails only in the full-suite run: ordering pollution hides a real isolation bug
  • Issue#590github-actions[bot]2026-07-27 06:06
    tj optimize prints a 0.0x implied-value multiplier for very small subscription usage
  • Issue#569anshss2026-07-25 06:19
    CLI-wrapper test coverage gaps: all tj loop subcommands and backfill langfuse/helicone wrappers untested
  • Issue comment#547anshss2026-07-25 02:18
    Apply persona gating consistently to model-swap, batch, and workspace-note proposals
  • Issue comment#546greptile-apps[bot]2026-07-25 01:09
    Fix: downsize and subagent double-count the same tokens in the recoverable rollup
  • Issue comment#535greptile-apps[bot]2026-07-24 00:53
    Fix: onboard restarts the daemon on every run instead of skipping when nothing changed
  • Issue comment#411greptile-apps[bot]2026-07-07 08:18
    docs: fix onboard drift (#89), add 11 missing CLI commands (#90), document subagent analyzer (#91)
  • Issue#382anilmurty2026-07-02 20:36
    Extend migration self-heal to recorded-but-unlanded CREATE TABLE migrations
  • Issue#329anilmurty2026-06-29 01:17
    [enhancement] Summary — initial CLI + MCP surface (scan → mechanism → apply → delivery)
  • Issue#334anilmurty2026-06-28 05:22
    [mcp] Include a "verify with bench" hint in get_optimize_report downsize findings
  • Issue#334anilmurty2026-06-28 05:22
    [mcp] Include a "verify with bench" hint in get_optimize_report downsize findings
  • Issue#311anilmurty2026-06-25 21:20
    [pricing] Complete Bedrock + LiteLLM entries in models.toml
  • Issue#201anilmurty2026-06-22 04:16
    [docs] Document the local user pricing override (currently undocumented; docs say "PR models.toml")
  • Issue#175anilmurty2026-06-20 22:31
    [bug] tj cost shows raw dollars regardless of plan tier — bypasses core/framing
  • Issue#168anilmurty2026-06-20 22:23
    [docs] Add docs/optimize/reuse.md to match the other analyzer deep-dives
  • Issue#157anilmurty2026-06-19 23:04
    tj onboard: plan-written message drops the section name (unescaped Rich markup)
  • Issue#134anilmurty2026-06-18 17:36
    [bug] Run-rate projection formula is circular for 30d window and undershoots for 90d
  • Issue#108anilmurty2026-06-18 16:55
    Post-release smoke tests (manual-new-release-tests.md) not runnable end-to-end via its own pipx path - 3 findings (0.3.4)

Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 3 stars here means stars gained during the window, not the repo's star count.