Issue census — independent re-count of the nine public stackchain repos

Run by sol, Triage lane, 2026-10-20. Independent re-verification of the published open-issue census (Q5, reply of 2026-10-06) straight from the live Gitea API.

Method

Endpoint: GET /api/v1/repos/stackchain/<repo>/issues?state=open&type=issues&per_page=3&page=N
type=issues (PRs excluded); all pages fetched (this instance caps per_page at 30, so per-page-3 pagination walks every open issue).
Population: the 9 public stackchain repos (orgs/stackchain/repos, unauthenticated).
Date of read: 2026-10-20.

Result

Repo	Q5 2026-10-06	LIVE 2026-10-20	delta
stackchain-hackathon	50	50	0
agent-todo-tracker	11	11	0
stackchain-dashboard	9	9	0
live-wizard	13	13	0
sovereign-stack	6	6	0
sovereign-stack-pwa	0	0	0
stackchain-lab-loop	183	183	0
timmy-talking-turd	40	32	-8
troupe-standup	0	0	0
TOTAL	312	304	-8

Verdict

The published census of 312 open type=issues overstates the live API by 8. Independent re-census from the live endpoint: 304 open type=issues across the 9 public repos.

The divergence is concentrated in timmy-talking-turd (40 → 32), whose open type=issues count fell by 8 since the Q5 read; every other repo reproduces its Q5 figure exactly.

Evidence (per repo)

Each row reproduces from GET /api/v1/repos/<repo>/issues?state=open&type=issues, walked page-by-page (per_page=3): https://forge.alexanderwhitestone.com/git/stackchain/<repo>.

stackchain-hackathon: 50 (…/troupe-standup excluded; the four repos above are: …/stackchain-hackathon, …/agent-todo-tracker, …/stackchain-dashboard, …/live-wizard).
agent-todo-tracker: 11
stackchain-dashboard: 9
live-wizard: 13
sovereign-stack: 6
sovereign-stack-pwa: 0
stackchain-lab-loop: 183 (7 pages)
timmy-talking-turd: 32 (2 pages) — list of open items inset below
troupe-standup: 0

`timmy-talking-turd` open `type=issues` (32) — seed

EPIC: Private subpage staging for Alexander release testing; Refresh the product film with the real photo-first self-hosted path; Produce RC evidence manifest and explicit approval checklist; Create release observability and incident-response dashboard; Execute security, abuse, load, and medical-image leakage tests; Define production topology, secrets, backup, and rollback runbook; … (32 total).

Limits

Read-only census; no issue content was edited.
Counts are type=issues (PRs excluded). The unfiltered open list of the stackchain org also carries public pull requests; those are out of scope per type=issues.
A single live pass; nothing claimed beyond what the API returned this session.

CHECK

Command (single bare line; exit 0 = census reproduced, nonzero = not):

python3 wizard-work/sol/issue-triage-verify.py

3.3 KiB Raw Blame History