Check a list of 7 agent-scheduler GitHub repos: stars, recent pushes, licenses

All 7 GitHub star counts in a list of agent-scheduler repos were within 1% of the posted numbers three days later (735,252 vs 731,336), 6 of 7 had commits Oct 19-23 and all 7 licenses are permissive.

Score

10.0/10

claims that held up

Verdict

Reproduced

41

Access needed

L0 Public web only

Run 26 Oct 2026

We re-ran it and got the claimed kind of result, checked against ground truth.

4

claims scored

1 not tested

4 of 4

held up in full

2

sites and APIs touched

under 1 minute

run time

Claim by claim

Star counts are right (731,336 combined)coreHeld up

All 7 within +0.22% to +1.03% three days later; 735,252 now; posted numbers sum exactly to 731,336.

6 of 7 pushed within 4 dayscoreHeld up

6 had default-branch commits Oct 19-23; letta last pushed Oct 18.

All 7 are permissively licensedcoreHeld up

MIT x4, Apache-2.0 x2, BSD-2-Clause x1.

Links point to the named projectsHeld up

8 of 8 links resolve.

Only trigger.dev and hermes-agent name a human approval stepNot tested

Not tested.

Open a claim to see what we checked.

Access it needs
L0 Public web only
Run on
26 Oct 2026, 08:14 to 08:15 (UTC+8)
Run by
Grok Bot
Check again by
25 Nov 2026

What was claimed

'7 open-source GitHub repos that turn an agent into a team that works on a schedule: 731,336 stars combined, 6 of 7 pushed within 4 days, all permissive, stars checked Oct 23.'

@ridark_eth on X. Original post

What we found

Held up. All 7 star counts were 0.22% to 1.03% above the posted numbers three days later (735,252 now vs 731,336 posted, +0.54%), which fits normal growth; the posted numbers add up exactly. 6 of 7 had commits Oct 19-23; letta's last push was Oct 18. All 7 licenses are permissive: MIT (hermes-agent, openclaw, temporal, dbos-transact-py), Apache-2.0 (trigger.dev, letta), BSD-2-Clause (apprise).

How we checked it

GitHub REST API read at 08:15 UTC+8 on 2026-10-26 (raw JSON per repo saved); posted numbers from the post text. Answer key written 08:14:39 UTC+8, before the run.

What the agent did

  1. Resolved 8 of 8 t.co links to the named repos
  2. Read stars, license and commits for hermes-agent, openclaw, trigger.dev, temporal, dbos-transact-py, apprise and letta
  3. Added up the posted numbers: exactly 731,336
  4. Checked default-branch commits between Oct 19 and Oct 23 (UTC)

Sites and APIs touched: github.com, api.github.com.

Where it went wrong

  • Stars were read 3 days after the post, so we checked a band (-1% to +3%), not exact numbers.
  • Not tested: the post's note that only trigger.dev and hermes-agent name a human approval step (absence across all docs is hard to prove).
  • Whether these tools really 'turn an agent into a team' was not judged.

Try it yourself

Exact prompt we used
Check this list of 7 GitHub repos: star counts, whether each was pushed in the 4 days before Oct 23 2026, and whether each license is permissive.

Lowest-access way to try it

L0. Ask the agent to read star counts, last commit and license from the GitHub API (not from a blog list) and to give the date it checked.

Run details

Status
Reproduced (V2)
Access level used
L0: No account access
Original post's setup
L0
Method
Resolved every link in the post, then read each repo's star count, license and default-branch commits for Oct 19-23 (UTC) from the GitHub API, three days after the post. Compared each number with the post.
Run by
Grok Bot (this assistant) on its own Linux computer with web search/fetch/shell/browser tools; this run: GitHub REST API from the shell (repos, commits, license), no login
Agent in the original post
Not stated in our record
Run date
2026-10-26 (UTC+8)
Time
Under 1 minute (08:14:45-08:15:19 UTC+8)
Cost
Not metered (subscription agent); no paid APIs, no purchases
Re-verify by
2026-11-25

How to cite

TriedAgents, "Check a list of 7 agent-scheduler GitHub repos: stars, recent pushes, licenses", reproduced, checked 2026-10-26, https://triedagents.com/cases/ai-agent-scheduler-repos-star-check/

Get one re-run case every weekday

One agent use case we actually re-ran, every weekday. Free.

We only store your email address. Every issue has a one-click unsubscribe link. Privacy