Release: version packages

[github-actions]

This PR was opened by the Changesets release GitHub action. When you're ready to do a release, you can merge this and the packages will be published to npm automatically. If you're not ready to do a release yet, that's fine, whenever you add more changesets to main, this PR will be updated.

Releases

@tangle-network/browser-agent-driver@0.36.1

Patch Changes

• #137 8199070 Thanks @AtelyPham! - Surface the agent's result in the CLI output and the bad view viewer.

  Previously, when a run's completion message was empty the agent's answer was written only to goalVerification.evidence in report.json; the CLI printed a bare "Goal achieved" and the viewer showed the same, so the actual result was hidden. Now:

  • The CLI prints the full result under the run's ✓/✗ line, with multi-line output preserved and indented (multi-task rows collapse to one line).
  • bad view renders it in a "Result" panel, derived directly from the report so it also shows for existing runs.

• #139 b7e9914 Thanks @drewstone! - Update the WebVoyager research command to use Agent Eval's maintained pairwise ranking API, and move the optional Claude Code provider to its secure AI SDK v6 release line.

[tangletools]

✅ Auto-approved GitHub Actions PR — 8ff752f0

This PR was opened by trusted GitHub Actions automation.
The full PR reviewer audit still runs separately and will publish findings if it detects issues.

This approval is provisional. It rests on the audit running. If the audit cannot run — for example the CLI bridge rejects it — this approval is dismissed rather than left standing, so an unrun check never reads as a passing one.

_{tangletools · auto-approval · reason: github_actions_author · 2026-08-13T19:23:40Z}

[tangletools]

🟢 Value Audit — sound

Verdict	sound
Coverage	2 of 2 lenses (value, usefulness)
Concerns	0 (none)
Heuristic	0.0s
Duplication	0.0s
Interrogation	33.5s (2 bridge agents)
Total	33.5s

💰 Value — sound

Automated Changesets release PR: consumes two patch changesets, bumps version 0.36.0→0.36.1, and publishes to npm on merge — does exactly what the repo's documented flow requires.

• What it does: A github-actions[bot] commit (8ff752f) that (1) deletes the two consumed changeset files .changeset/print-agent-result.md and .changeset/update-agent-eval.md, (2) appends a 0.36.1 section to CHANGELOG.md containing their summaries verbatim, and (3) bumps package.json version from 0.36.0 to 0.36.1. It is the release vehicle for PRs #137 (surface agent result in CLI/bad view) and #139 (WebVoyager →
• Goals it achieves: Cut a patch release publishing @tangle-network/browser-agent-driver@0.36.1 to npm via the documented Changesets+OIDC pipeline. Merging it is the only step that triggers publish — the underlying code changes are already on main.
• Assessment: Correct and fully in-grain. CLAUDE.md documents this exact flow ('PR merges to main → changesets.yml opens a Release PR bumping package.json + CHANGELOG.md → merging that PR publishes via OIDC'). Both source changesets were patch-level, so the patch bump 0.36.0→0.36.1 is right. The bot faithfully moved the changeset text into CHANGELOG without alteration, matching the 'summaries land verbatim' rul
• Better / existing approach: none — this is the right approach. The repo deliberately uses Changesets as its release mechanism (.github/workflows/changesets.yml + .changeset/config.json), and this PR is the exact output that mechanism is designed to produce. A manual version bump or a different release tool would fight the established grain. No existing equivalent to 'reinvent' — this IS the mechanism.
• Model: opencode/zai-coding-plan/glm-5.2
• Bridge attempts: 2
• Bridge warning: opencode/kimi-for-coding/k2p7: opencode: opencode error

🎯 Usefulness — sound

Routine, fully-automated Changesets release PR: bumps package.json 0.36.0 → 0.36.1 and regenerates CHANGELOG from two consumed patch changesets; merging is what triggers npm publish.

• Integration: Wired correctly into the repo's documented release path. .github/workflows/changesets.yml:71-114 consumes this exact commit: on merge to main it detects no leftover changesets, runs pnpm release-tag to push the vX.Y.Z tags, then dispatches publish-npm.yml (changesets.yml:107-114) for OIDC npm publishing. The version field in package.json is the single input the publish step reads (changesets.yml
• Fit with existing patterns: This is THE established release pattern, not a competing one. CLAUDE.md documents the Changesets + OIDC flow as canonical; git log shows ~8 prior 'Release: version packages' PRs (#128, #132, #134, #136...). Both consumed changesets (.changeset/print-agent-result.md, .changeset/update-agent-eval.md) declare patch and map verbatim to merged PRs #137 (fix(cli)) and #139 (fix(eval)); the 0.36.0 → 0.
• Real-world viability: Holds up: there is no happy/sad path to break — the change is two declarative fields (version string, changelog prose) generated deterministically by changesets/action@v1 from already-merged, already-gated PRs. The underlying code (#137, #139) was reviewed and signed off independently; this PR carries no logic. No error paths, no concurrency surface beyond the workflow's own concurrency group (cha
• Model: opencode/zai-coding-plan/glm-5.2
• Bridge attempts: 1

No concerns — sound change, no better or existing approach found. ✅

---

What this audit checks

It judges the change on its merits — not whether it was tasked out in an issue. Unticketed, fast-moving work is fine; the question is whether the change is good and whether a better or existing approach should be used instead.

Pass	What it asks
Heuristic	Vague title? Whitespace-only or cruft-bearing diff? (content signals only)
Duplication	Do added function/class names already exist elsewhere in the repo?
Value Audit	What does it do? What goal does it achieve? Is it good? Better architecture or already-exists?
Usefulness Audit	Does it integrate and fit? Will it hold up in real use and actually get used?

Findings are concerns, not blocks — the human reviewer decides what to do with them.

_{value-audit · 20260813T195405Z}

[tangletools]

✅ No Blockers — 8ff752f0

Review health 100/100 · Reviewer score 95/100 · Confidence 70/100 · 0 findings (none)

	opencode GLM 5.2	opencode DeepSeek v4 Pro	opencode DeepSeek v4 Flash	aggregate
Readiness	95	95	95	95
Confidence	70	70	70	70
Correctness	95	95	95	95
Security	95	95	95	95
Testing	95	95	95	95
Architecture	95	95	95	95

Reviewer score is advisory once the run is complete and the verdict has no blockers.

Full multi-shot audit completed 2/2 planned shots over 2 changed files. Global verifier still owns final merge decision. | Full multi-shot audit completed 2/2 planned shots over 2 changed files. Global verifier still owns final merge decision. | Full multi-shot audit completed 2/2 planned shots over 2 changed files. Global verifier still owns final merge decision.

No findings.

---

_{tangletools · 2026-08-13T19:54:34Z · trace}

[tangletools]

✅ Clean — 8ff752f0

Full multi-shot audit completed 2/2 planned shots over 2 changed files. Global verifier still owns final merge decision. | Full multi-shot audit completed 2/2 planned shots over 2 changed files. Global verifier still owns final merge decision. | Full multi-shot audit completed 2/2 planned shots over 2 changed files. Global verifier still owns final merge decision.

Full immutable report for this review: trace

Summary comment for this run: full summary

---

_{tangletools · 2026-08-13T19:54:34Z · immutable trace}