release: v1.34.0

[Desperado]

Release v1.34.0.

Ships the shared conversation context work merged in #192 — the only change since v1.33.0.

Highlights

• Context survives switching models, effort levels, and providers. A lossless portable transcript is retained separately from the compacted working context; CLI backends reuse their native session and receive only the turns they missed.
• /save, /resume, --resume on every backend, one-shot -p included. Large conversations stay reachable via a private searchable JSONL archive.
• Conservative redaction for retained content — a saved transcript is replayed as the conversation, so only unambiguous credential shapes are rewritten. Structured fields still redacted by key name.
• Bounded session storage — 90 days for portable transcripts, 7 for legacy; no more double-storing the same conversation.
• Fixes — turn counts exclude tool traffic, archive cleanup on Ctrl+C/SIGTERM, unwritable temp dir no longer fails a turn, OpenCode nested tool state keeps its input.

Minor bump per SemVer: new backward-compatible functionality, no breaking changes to flags, config, or saved-session compatibility (older session files still load).

Merging this and pushing the v1.34.0 tag triggers release.yml, which builds the six OS/arch binaries and publishes the GitHub release.

[qualitymaxapp]

QualityMax Review

Verdict: COMMENT · Confidence: unavailable — no completed model review

Files eligible: 1 · Files reviewed: 1 · Files with findings: 0 · Findings: 0 · Inline cards: 0

Priority findings

priority	location	finding
—	—	No blocking findings

Review gates

gate	status
AI diff review	skipped · eligible 0, reviewed 0 · LLM · All changed files were excluded as non-code, generated, or binary content.
SAST	completed · eligible 1, reviewed 1 · hybrid
Overall review evidence	unavailable — insufficient model review evidence
Inline evidence	not needed

Important files

file	risk	note	next step
—	—	No findings	—

Review lifecycle

Use the inline cards to inspect evidence and suggested remediation. Re-run the QualityMax review after pushing a fix; unchanged cards are identified by their stable finding marker. Dismiss with a reason through the existing QualityMax/GitHub review feedback flow. 0 prior card(s) are stale/resolved on this head. @qmax Q&A is tracked separately.

Proof legend: VERIFIED independently judged patch · REPRODUCED verified finding · GROUNDED deterministic evidence · MODEL-ONLY model judgment.

QualityMax project results are available in the configured project.

_{Receipt · commit 515c7c9202bba0ed675f737759da047be3071097 · run 2026-09-08T12:35:16+00:00 · model served not applicable · model requested not applicable · model review unavailable — no completed model review · model source not recorded · re-review 2 · proof counts {}}

[qualitymaxapp]

QualityMax Review — canonical overview updated; inline findings are attached to this review.

[qualitymaxapp]

⚠️ QualityMax Pipeline

Gate	Result
🔍 AI diff review	⏭️ Skipped — All changed files were excluded as non-code, generated, or binary content. · skipped · 0 eligible / 0 reviewed · All changed files were excluded as non-code, generated, or binary content.
🔍 SAST	completed · 1 eligible / 1 reviewed
🔍 Canonical PR review delivery	completed · 0 eligible / 0 reviewed · exact-head review #5141723863 and overview #5585214636 confirmed
🧪 Repo Tests	✅ 794/794 passed (go)

_{Powered by QualityMax — AI-Powered Test Automation}