[good-first-issue scan] 2026-08-17

[vivekchand]

Today's candidates (3)

clawmetry-cloud#321: [P1] Compaction markers with expandable summary

Scope: Cloud session transcript view - routes/cloud.py session endpoint (~20-30 LOC for compaction event type) and session JS/HTML template (collapsible divider + expandable markdown card, ~50 LOC). All within the session-transcript module. OSS reference implementation: merged PR clawmetry#662.
Why safe: No P0 in title; [P1] prefix is not in any disqualified-prefix list; no disqualifying labels; no epic; does not touch auth, billing, Stripe, Cloud Run config, CI workflows, sync daemon, proxy, or gateway WebSocket; no open PRs in clawmetry-cloud; last comment 2026-04-14 (well outside 14-day window, no bot or claim signals); clear done state - compaction divider renders in transcript, click expands full markdown summary.
Apply command: gh issue edit 321 --repo vivekchand/clawmetry-cloud --add-label "good first issue"

---

clawmetry-cloud#320: [P1] Model-change + thinking-level-change annotations on session timeline

Scope: Cloud session timeline and sidebar - expose model_change/thinking_level_change event types in the session endpoint (~30-40 LOC) and session JS/HTML templates for timeline stripe markers plus "Model journey" sidebar panel (~100 LOC). All within the session-detail module. OSS reference implementation: merged PR clawmetry#753.
Why safe: No P0 in title; [P1] prefix not disqualified; no disqualifying labels; no epic; no auth/billing/Stripe/Cloud Run/CI/sync/proxy/gateway/security touch; no open PRs in clawmetry-cloud; last comment 2026-04-14 (no bot or claim signals in last 14 days); clear done state - timeline shows model-switch stripe annotations, sidebar shows ordered model list with duration and cost-per-model.
Apply command: gh issue edit 320 --repo vivekchand/clawmetry-cloud --add-label "good first issue"

---

clawmetry-landing#586: Copy sweep: 15 runtimes (n8n shipped) - hero counts, tooltips, how-it-works, llms.txt

Scope: Landing site copy files only - index.html (hero tooltip list), en.json (runtime count strings), how-it-works template, control_tower.hero copy, llms.txt, llms-full.txt. Purely additive string replacements, ~40-60 LOC total. All within the clawmetry-landing site module.
Why safe: No P0 in title; no disqualifying prefix; no labels at all; no epic; no auth/billing/Stripe/CI/security touch; only open PR (#645) touches blog files, does not reference #586; 0 comments (no bot or claim signals); clear done state - searching the repo for the old runtime count strings returns zero results.
Apply command: gh issue edit 586 --repo vivekchand/clawmetry-landing --add-label "good first issue"

---

Rejected close calls (3)

• clawmetry#2258 "i18n pseudolocale (en-XA) audit pass": otherwise the strongest OSS candidate (i18n.js + locale catalog, crystal-clear done state) but already carries the good first issue label - excluded by label rule.
• clawmetry-cloud#322 "[P2] Full conversation replay with time scrubber": passes all disqualification rules but scope spans a horizontal scrubber component, timeline filter logic, a live state panel, and a play/pause timer - too wide for single-module, and the OSS-iframe architecture makes it cross-repo.
• clawmetry-landing#375 "chore(seo): submit sitemap to Google + Bing and validate Rich Results": passes all rules but produces zero code changes - the entire task is external console submissions with no verifiable repo diff, so "done" is invisible to a stranger reading a PR.