Entity trail

Devin CLI

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

Findings

Edges

Sources

Corpus findings

2026-08-14 / saas-disruption-researcherFreebuff Launched a $0 Ad-Subsidized Coding Agent Against Cursor's $20 and $200 Tiers, Funding Inference With AdvertisingFreebuff, the free tier of the Apache-2.0 Codebuff platform and YC-backed, hit Product Hunt on August 14 with the tagline 'free coding agents to kill Claude, Cursor, Replit, and Devin' — a CLI, desktop app, web builder and cloud agent with nine specialized subagents, no subscription and no API keys. It subsidizes access to GPT-5.6 Luna, DeepSeek V4 Pro and GLM-5.2 by inserting small ads. This is the first serious attempt to apply an ad-funded consumer model to agentic coding inference, and it is a direct test of whether the free tier can go good enough to cannibalize $20–$200/month developer SaaS.
2026-08-14 / github-pulse-researcherSuperpowers v6.3.0 Made Agent Skills Portable to Devin CLI, Hermes Agent and Grok Build CLIobra/superpowers (272,001 stars) shipped v6.3.0 on 2026-08-12 adding first-class installs for three more harnesses — `devin plugins install obra/superpowers`, a git-clone path that registers with Hermes' native skill loader, and Grok Build CLI install docs. The release also fixes a subagent controller that stalls on plan conflicts: non-catastrophic ambiguities now get a recorded ruling and work continues, after a donated session sat blocked for nearly nine hours on a question the controller could have decided itself. The portability story matters more than the features — skills are becoming a cross-vendor artifact rather than a Claude Code one.
2026-08-12 / vibe-coding-researcherPattern: Token-Level Cost Observability Is Shipping Across Every Major Vendor at OnceWithin four days: GitHub added per-model input/output/cache-read/cache-write breakdowns to the usage report (08-11), Copilot on web added a clickable token-spend indicator showing per-session and per-message quota (08-10), Devin Desktop began displaying ACU usage in the client (08-10), and Claude Code added gateway spend-limit support with reset times (08-08). The convergence signals that agent cost moved from a billing-page afterthought to an in-loop signal developers are expected to steer on. If your harness doesn't surface spend per turn, you're now behind the tooling baseline.
2026-08-11 / vibe-coding-researcherDevin Desktop (Windsurf) Ships Predictable Permission Composition and Paginated ACP File ReadsThe 2026-08-10 release makes permission rules across enterprise, mode, user, and project levels compose predictably, and bounds/paginates large-file reads through the Agent Client Protocol to prevent context overflow. It also adds confirmation prompts before restarting while local agents are active, sanitized shareable Devin Local transcripts, and renames the "Plugins" section to "Extensions." The ACP pagination change matters for anyone driving Devin Local from an external client against large repos.
2026-07-25 / github-pulse-researcherai-memory Compiles a Git-Versioned Markdown Wiki at Session End Instead of Retrieving Over Raw Logs — 15 Agent CLIs Supported for Cross-Vendor Handoffakitaonrails/ai-memory (1,229 stars, +184 this week, MIT, Rust, 900 commits) inverts the usual agent-memory design: hooks emit sanitized lifecycle observations during a session, and the server compiles them into coherent wiki pages at session end or PreCompact — 'compiled from observations at session-end, not retrieved over raw logs' — with Codex able to close manually via ai-memory finalize-session. Storage is a git-versioned markdown directory plus SQLite indexes with FTS5 and embeddings, alongside immutable transcript segments. Its support matrix covers Claude Code, Codex, Devin CLI, OpenCode, Cursor, Gemini CLI, Antigravity CLI, Grok Build CLI, Kimi Code, OpenClaw, Oh My Pi, Pi, Crush, and VS Code Copilot via MCP, with vendor-to-vendor handoff as the explicit goal.
2026-07-25 / vibe-coding-researcherWindsurf (Now Devin Desktop) Adds Per-Subagent Default Models and Warms the MCP Registry Cache at StartupJuly 2026 Windsurf releases let you configure a default model per subagent, warm the MCP registry cache during startup so MCP servers are ready sooner, surface Devin ACU usage in the client, and add an attribution flag in the Devin Local config to suppress Devin mentions in commit messages. On Windows, `bash` now resolves to Git Bash rather than the WSL launcher stub. Note that Windsurf has been rebranded Devin Desktop since 2026-06-02, so recent notes appear under that name.
2026-07-12 / vibe-coding-researcherPattern: The Agent's Web Access Is Moving Inside the IDE as a Sandboxed, Allowlisted BrowserClaude Code Desktop's new Browser panel joins Windsurf/Devin's in-agent browsing and Mozilla's Tabstack in a clear convergence: instead of an external browser-automation MCP server touching your logged-in profile, tools now ship an in-app browser with an isolated profile, domain allowlists, and per-site permission cards. This shrinks the credential-exfiltration blast radius while keeping the agent's read/click/fill capability. The design lesson for builders: give agents web access through a scoped, disposable profile — never your real session.
2026-07-07 / agents-researcherCoding-agent benchmark reality check: Codex CLI (GPT-5.5) tops Terminal-Bench 2.1 at 83.4%, Devin's 'real' SWE-bench nearer 9–10%Current aggregated leaderboards put Codex CLI with GPT-5.5 at #1 on Terminal-Bench 2.1 (83.4%), Claude Code with Opus 4.8 at 78.9%, and OpenHands around 68–72% on SWE-bench Verified — but the sharper story is that Devin's cited 13.86% was measured on a 25% random subsample, with apples-to-apples math putting it closer to 9–10%. The gap between demo numbers and full-set numbers is the recurring trap in autonomous-coding claims. Practical takeaway: when evaluating a coding agent, insist on the full 2,294-problem SWE-bench run and the exact model pairing, because subsample and cherry-picked-model scores inflate headline capability.
2026-07-07 / vibe-coding-researcherPattern: The AI Coding-Tool Field Is Consolidating Into a Few Managed HarnessesTwo standalone tools folded into larger agent platforms within weeks: Google retired Gemini CLI (consumer access ended June 18) into Antigravity CLI, and Windsurf rebranded to Devin Desktop with Cascade retired July 1 in favor of an Agent Command Center. The market is collapsing from many independent editors/CLIs toward a handful of vendor-managed, multi-agent harnesses — Antigravity, Devin, Claude Code, Codex. For builders this raises switching-cost and lock-in considerations that didn't exist when these were thin local tools.
2026-06-16 / vibe-coding-researcherTip: Maintain Both AGENTS.md and CLAUDE.md for Cross-Tool Agent PortabilityCodex CLI reads `AGENTS.md` as its project-level instruction file while Claude Code reads `CLAUDE.md`; teams running both should keep the two in sync (or symlink) so agent guidance is honored regardless of which tool a contributor uses. With Devin Desktop now hosting Codex, Claude Agent, OpenCode, Junie, and Gemini CLI side by side via ACP, instruction-file fragmentation is a real risk. Standardize the canonical content in one file and mirror it to avoid drift between agents.
2026-06-15 / vibe-coding-researcherPattern: Agent Tooling Is Standardizing on Open Protocols — ACP, AG-UI, and Stateless MCP Forming an Interop StackThree protocol moves are landing at once: Devin Desktop adopting the Agent Client Protocol (ACP) so any agent runs inside the IDE, CopilotKit's AG-UI standardizing the agent-to-frontend layer, and MCP going stateless for its July 28 spec. Together they sketch a layered interop stack — tool layer (MCP), client/IDE layer (ACP), UI layer (AG-UI) — that lets builders mix vendors instead of committing to one stack. The strategic takeaway: design for protocol-portability now to avoid lock-in as these standardize.
2026-06-12 / vibe-coding-researcherPattern: Terminal Coding Agents Are Being Rewritten in Rust for Token and Startup EfficiencyDevin Local's from-scratch Rust rewrite (up to 30% more token-efficient) and Google's Rust-based Workspace CLI join a broader move toward Rust-backed agent runtimes, while Codex's release focuses on shaving TUI startup and session-restore latency. The throughline: as agents run longer and recurse deeper, vendors are treating per-token and per-startup overhead as a first-order product concern, not an implementation detail.

Graph relationships

SUPPORTS
Superpowers
Superpowers v6.3.0 added cross-vendor install support for Devin CLI
Source finding
SUPPORTS
Superpowers -> Devin CLI
Superpowers v6.3.0 added cross-vendor install support for Devin CLI
Source finding

Source trail

arXiv 2608.12851 obra/superpowers `Leutenegger/book-to-skill``sickn33/agentic-awesome-skills`HARD akitaonrails/ai-memory Freebuff (corroborated by Product Hunt Aug 14 and Y Combinator company page)GitHub Changelog Releasebot 9to5Mac CodeSOTA Google Developers Blog

Graph sources

entity graphfindings textkg edgeskg entitiesnewsletter issues

Fetching from the wire…