Skip to content
Guidance/Daily Briefing
digesteveryonedigestevidenceradarmodelagent6 min read · Aug 10, 2026

Radar brief — 2026-08-10

Codex and Claude push plugin/self-host surfaces hard; Cline hardens checkpoints and MCP OAuth; security patches land in Codex, Gemini CLI, and Windsurf

What the evidence shows

High-signal versioned releases cluster around agent plugins, session/checkpoint control, MCP auth, and safer defaults—not raw model swaps.

OpenAI Codex CLIrust-v0.147.0
Key Signal & Impact
Portable Agent Plugins + catalog search; conversation sections; --approve-for-me; import Cursor-managed skills and sync Claude/Cursor threads without dupes
OpenAI Codex CLIrust-v0.146.0
Key Signal & Impact
/new//clear session naming, pin/side threads; Agent Plugins manifests & marketplaces (Bedrock, Claude Code); thread forks; app-server ↔ remote Code Mode over WebSocket
OpenAI Codex CLIrust-v0.146.1
Key Signal & Impact
Safer automatic-review defaults for cyber-capable models; permission changes explained in TUI
Claude Codev2.1.224
Key Signal & Impact
claude self-hosted-runner for Team/Enterprise; archive plugin source (zip + optional SHA-256); ANTHROPIC_BEDROCK_REGION_PREFIX; cancel-and-confirm on paste removal
Clinedesktop-v0.0.8
Key Signal & Impact
Edit earlier messages → fork + workspace checkpoint rewind; transactional/atomic restores; more reliable checkpoints (restart/compaction/resume)
Clinev4.1.7
Key Signal & Impact
“View Changes” via SDK checkpoints; pre-registered OAuth clients for remote MCP; centralized plugin settings/toggles
Clinesdk/sdk/v0.0.72
Key Signal & Impact
Queued prompts survive aborts; durable session context across aborts/hub restarts; failed queued turns → run.failed
Google Antigravity1.1.10
Key Signal & Impact
Business sign-in for Gemini Enterprise (Cloud project, license seats, regional inference, admin controls)
Goosev1.45.0
Key Signal & Impact
Stable agent event IDs; model tab completion/provider switch; latest Gemini support; GOOSE_DOCS_ROOT for air-gapped docs; templated compaction summaries
OpenCodev1.18.5
Key Signal & Impact
Claude adaptive-thinking fixes; Mistral reasoning/cache stability; prompt-cache key correctness; terminal/server discovery fixes
Gemini CLIv0.54.0
Key Signal & Impact
HTTPS enforced for GoogleCredentialsAuthProvider; session ID rotate on model fallback; caretaker triage hardening (also in preview/nightly lines)
Windsurfv3.6.27
Key Signal & Impact
Windows cert-store fix for TLS-inspecting proxies; local edit tools refuse writes through symlinks
Zed AIv1.15.0-pre
Key Signal & Impact
git.diff_base setting; Project Panel drag-out; JSX/TSX linked editing / Emmet in bodies (prerelease)
xAI Grok CLI (Grok Build)8a14c91
Key Signal & Impact
Worktree/resume flag behavior; skill-path suggestions; bounded .envrc eval; tools-server OOM protection (commit sync, not a semver tag)

Confidence note: Gemini CLI preview/nightly entries overlap changelogs and lean “caretaker”/internal workflow—noisier than the stable v0.54.0 security fixes. Windsurf is changelog HTML (weaker than GitHub release tags). Grok is a monorepo commit hash.

For developers

  • Codex CLI: Try rust-v0.147.0 if you care about plugin catalogs, --approve-for-me, or Cursor/Claude skill import; read permission/auto-review notes from rust-v0.146.1 before wide --approve use.
  • Claude Code: On Team/Enterprise, evaluate self-hosted-runner and pin plugins via archive + SHA-256 when git/npm are blocked.
  • Cline: Use desktop-v0.0.8 when you need mid-thread edit → full workspace rewind; use v4.1.7 “View Changes” + remote MCP OAuth for reviewable agent diffs.
  • Goose: v1.45.0 if you need air-gapped docs root, Gemini model picks, or stabler compaction summaries.
  • OpenCode / Gemini CLI / Windsurf: Pull current patches for thinking/cache correctness, HTTPS credential transport, and symlink-write refusal—these are foot-gun fixes, not features.
  • Zed / Grok / Gemini nightlies: Treat as exploratory only until non-pre releases cover your workflow.

For teams

  • Policy & approvals: Codex’s --approve-for-me plus cyber-model auto-review defaults need an explicit team policy (who may enable, which repos, audit trail).
  • Enterprise identity: Antigravity 1.1.10 Business sign-in and Claude self-hosted runners are the org-control levers in this window—map them to existing IdP, region, and seat processes before rollout.
  • Supply chain: Prefer Claude archive plugins with SHA-256 pins and Codex workspace/plugin marketplace controls over ad-hoc installs.
  • MCP: Standardize on Cline’s pre-registered OAuth path where dynamic client registration is blocked.
  • Security baselines: Require Windsurf ≥ v3.6.27 (symlink write block + corp CA), Gemini CLI ≥ v0.54.0 (HTTPS auth provider), and Codex ≥ rust-v0.146.1 (safer cyber-model review) in managed images.
  • Checkpoint culture: Cline’s atomic restore/checkpoint work is the concrete enablement for “edit history safely” training—pair with mandatory review of completion-card diffs.

What to ignore

  • Marketing tone and unversioned “AI coding” claims outside this evidence set.
  • Gemini caretaker/Firestore/Pub/Sub orchestration noise in nightlies unless you run that internal stack.
  • Zed v1.15.0-pre and Gemini nightlies/previews as production pins.
  • Open proposals still at P5 backlog (older Codex/Cline/Goose/OpenCode/Zed items)—they are not new release proof.
  • Inflated “+N more changes” tails without reading the tagged notes that matter for your surface (MCP, sandbox, approvals).

Watch next

  • Codex plugin marketplace + remote Code Mode adoption vs. breakages tagged breaking/surface:mcp.
  • Claude self-hosted-runner operational patterns (network, secrets, plan gating).
  • Whether Cline checkpoint/OAuth work lands consistently across desktop, extension, and SDK hosts.
  • Follow-up stable Zed cut after v1.15.0-pre.
  • Grok Build moving from commit syncs to clearer versioned releases.
  • Further cyber-model / approval default tightening after Codex rust-v0.146.1.

How to use DevAgentRadar

  • Read What the evidence shows as the only source of truth for versions; don’t infer unlisted releases.
  • Developers: pick one assistant version and one concrete action (plugin pin, checkpoint try, security bump).
  • Teams: turn the same rows into allowlists, image pins, and approval policy—not slide decks.
  • Ignore low-trust or prerelease rows until they reappear as stable tags.
  • Re-check open proposals and the next brief before mandating fleet-wide upgrades.

Evidence appendix

  • Evidence window: top curated signals (versioned releases preferred)
  • Assistants tracked: 22
  • Signals in this brief: 16
  • Trusted source mix (GitHub/changelog vs HTML): 55%
  • Open improvement proposals: 8
  • Sources reporting errors: 0

High-signal releases used

  • OpenAI Codex CLI rust-v0.146.0: Name new sessions with /new or /clear, pin important threads, and switch between side conversations without closing them. (#34605, #34840, #35011) · Support — source

  • Claude Code v2.1.224: Added self-hosted environments: claude self-hosted-runner turns your own machines or containers into a place Claude Code web, mobile, and desktop sessions can — source

  • OpenAI Codex CLI rust-v0.147.0: Install portable Agent Plugins and search across local, personal, workspace, and remote plugin catalogs. (#36544, #36409, #36919, #36796) · Organize conversatio — source

  • Zed AI v1.15.0-pre: This week's release includes a new git.diff_base setting for choosing whether editor gutters, file-status colors, and git: diff show uncommitted changes aga — source

  • Google Antigravity 1.1.10: Added Business sign-in for Gemini Enterprise accounts, so you can authenticate with a Google Cloud project under Google Cloud terms, use a license seat allocate — source

  • Cline desktop-v0.0.8: Edit any earlier message in a conversation — the app forks the session at that point, rewinds the workspace to that run's checkpoint, and re-runs from your edit — source

  • Cline v4.1.7: Restore the "View Changes" button on completion rows, backed by SDK checkpoints, so you can review everything a task touched from the completion card. · Bring b — source

  • Cline sdk/sdk/v0.0.72 — SDK v0.0.72: Prompts queued during a turn now survive being interrupted: they are preserved across user-initiated aborts, drained after a turn aborts itself, and edits made — source

  • OpenAI Codex CLI rust-v0.146.1: Apply safer automatic-review defaults for cyber-capable models and explain permission changes in the terminal interface. (#37057) · #37057 [0.146] Backport safe — source

  • Goose v1.45.0: Stable agent event message identity #10716 · Tab completion and switch provider for model command [#10585](h — source

  • Gemini CLI v0.56.0-nightly.20260808.gcf22ac7e8: Reclassifying Capacity Exhaustion as Terminal Error · feat(caretaker): update Firestore schema with error, and pr_number fields · feat(caretaker-triage): prompt — source

  • OpenCode v1.18.5: Improve Claude adaptive thinking handling across more response shapes. · Avoid OpenAI Responses phase handling that could break some conversations. · Preserve g — source

  • Gemini CLI v0.54.0-preview.0: Changelog for v0.53.0-preview.0 · Changelog for v0.52.0 · chore(release): bump version to 0.54.0-nightly.20260722.gf743ab579 · fix(caretaker): sanitize and wrap — source

  • Gemini CLI v0.54.0: Changelog for v0.53.0-preview.0 · Changelog for v0.52.0 · chore(release): bump version to 0.54.0-nightly.20260722.gf743ab579 · fix(caretaker): sanitize and wrap — source

Open proposals

Was this guidance useful?

Anonymous · one vote per visitor · re-click to clear. Helps improve digests—not a product ranking.