fix(delegation): reuse native agent monitoring presentation

[rynfar]

Pylon delegation added a separate summary above the chat and boxed child cards instead of fitting the existing native-agent interface. Remove that summary and share the native streaming CTA and three-line Agents-panel row components.

Delegation spawn entries now use the same completed-turn and tool-grouping exemptions as native spawn entries, so collapsing the parent's work log does not hide them. Selecting an entry opens the existing Agents panel; tool diagnostics remain expandable, and the panel is the only roster. The existing active-agent badge continues to include child shells after the parent stops and describes approval/input blockers as needing attention.

Child status is matched only by exact thread ID. Truncated results retain a neutral recorded label and route to the authoritative panel; no title matching or borrowed child status. Providers without structured tool identity retain generic logs and use the existing badge/panel. Native controls and token totals are unchanged. No new backend polling or paid provider runs.

Validation: 182 focused tests across timeline rendering/grouping, panel rows, and badge states; web typecheck; scoped lint/formatting; whitespace checks. Independent adversarial review found no blocking issues and independently ran the affected tests. Regression coverage includes completed parents with active/blocked/completed children, truncated results with multiple unrelated children, and truthful waiting states. Browser verification used the earlier disposable smoke-test conversation: collapsed work log retains the entry, clicking opens Agents, and the separate top summary is absent.

Before: #593 (comment)

Implemented with GPT-6 Astra in Codex, with built-in implementation and independent adversarial-review agents.

---

^{Need help on this PR? Tag @codesmith-bot with what you need. Autofix is disabled.}

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated
pylon-marketing	Ready	Preview	Sep 17, 2026 5:14pm UTC

[rynfar]

After: the existing streaming entry opens the existing Agents panel. No separate summary above chat and no inline child roster. This screenshot uses the prior completed smoke-test conversation; its historically truncated MCP result intentionally shows a neutral recorded label, while the panel shows the actual child completion.

[github-actions]

Thread transfer impact

✅ Thread transfer remains within every enforced ceiling.

Provider	Metric	Main baseline	This PR	Impact	PR ceiling
Codex	Total thread wire	13.9 KiB	14.0 KiB	+5 B (+0.0%)	15.1 KiB	✅
Codex	Thread snapshot wire	7.2 KiB	7.2 KiB	−7 B (−0.1%)	7.3 KiB	✅
Codex	Live turn WebSocket wire	6.7 KiB	6.7 KiB	+12 B (+0.2%)	7.8 KiB	✅
Codex	Live turn WebSocket decoded	58.0 KiB	58.0 KiB	0 B (0.0%)	66.4 KiB	✅
Codex	Live turn messages	9	9	0 (0.0%)	21	✅
Claude	Total thread wire	13.9 KiB	14.0 KiB	+45 B (+0.3%)	15.1 KiB	✅
Claude	Thread snapshot wire	7.2 KiB	7.2 KiB	+9 B (+0.1%)	7.3 KiB	✅
Claude	Live turn WebSocket wire	6.7 KiB	6.8 KiB	+36 B (+0.5%)	7.8 KiB	✅
Claude	Live turn WebSocket decoded	58.8 KiB	58.9 KiB	+44 B (+0.1%)	66.4 KiB	✅
Claude	Live turn messages	8	9	+1 (+12.5%)	21	✅

Baseline: 81975aa · PR result: f1f626a · Source CI: success

Scenario and decoded snapshot size

10 historical turns, 5 command tools per turn, 878.9 KiB retained MCP result per historical turn, and a 1.05 MiB retained result in the measured turn.

• Codex decoded thread snapshot: 115.6 KiB
• Claude decoded thread snapshot: 116.3 KiB

Updated in place by a trusted workflow. PR artifacts are strictly validated and never executed.