diff --git a/newsletters/2026-March.md b/newsletters/2026-March.md
new file mode 100644
index 0000000000..1d42443dcd
--- /dev/null
+++ b/newsletters/2026-March.md
@@ -0,0 +1,86 @@
+# Spec Kit - March 2026 Newsletter
+
+This edition covers Spec Kit activity in March 2026. Versions v0.2.0 through v0.4.3 shipped during the month — nine releases — introducing major capabilities including simultaneous multi-catalog extension support, a pluggable preset system, air-gapped offline deployment, and automatic skill registration for extensions. Seven new AI coding assistants were integrated, bringing total platform support past 22. Community activity included over twenty new extensions, independent walkthroughs and blog posts, and a wave of industry coverage debating whether "vibe coding" is dead. A category summary is in the table below, followed by details.
+
+| **Spec Kit Core (Mar 2026)** | **Community & Content** | **SDD Ecosystem & Next** |
+| --- | --- | --- |
+| Versions **v0.2.0** through **v0.4.3** shipped with major features: multi-catalog extensions, pluggable presets, air-gapped deployment, and auto-registration of extension skills. Seven new agents added (Tabnine CLI, Kimi Code, Mistral Vibe, Junie, iFlow, Trae, Pi). The repo grew from ~71k to **72,700 stars** by March 20. [\[github.com\]](https://github.com/github/spec-kit/releases) | Walkthroughs by Tiago Valverde, Alfredo Perez, and Sergey Golubev covered SDD in practice. Over 20 community extensions reached the catalog. The Spec Kit Assistant VS Code extension was recognized as a Community Friend. A Microsoft Learn training module on SDD with Spec Kit was available. [\[tiagovalverde.com\]](https://www.tiagovalverde.com/posts/spec-driven-development-in-practice-a-walkthrough-with-spec-kit) [\[alfredo-perez.dev\]](https://www.alfredo-perez.dev/blog/2026-03-21-build-your-own-sdd-workflow) | ByteIota reported AWS pushing SDD as the new standard; multiple independent articles declared "vibe coding" dead. Augment Code published a detailed Spec Kit vs. Intent comparison. Competitors differentiate on orchestration depth and living specs; Spec Kit leads in agent breadth and portability. [\[byteiota.com\]](https://byteiota.com/spec-driven-development-kills-vibe-coding-march-2026/) [\[augmentcode.com\]](https://www.augmentcode.com/tools/intent-vs-github) |
+
+***
+
+## Spec Kit Project Updates
+
+### Three Major Versions and Six Patches
+
+**v0.2.0** (released March 10) was the month's opening milestone. Its headline feature was **simultaneous multi-catalog support** (PR #1720), enabling users to activate both the core and community extension catalogs at the same time — a prerequisite for the modular ecosystem that would flourish throughout the rest of the month. The release bundled the first new agent integrations of March — **Tabnine CLI** (#1503) and **Kimi Code CLI** (#1790) — along with four community extensions: **Understanding** (#1778), **Ralph** (#1780), **Review** (#1775), and **Fleet Orchestrator** (#1771). Tooling improvements included `.extensionignore` support for excluding files during extension installation (#1781) and **Codex extension command registration** (#1767). Patch **v0.2.1** followed to fix broken quickstart links (#1759/#1797), add catalog CLI help documentation (#1793/#1794), and use quiet checkout to suppress git exceptions (#1792). The February 2026 newsletter was also committed as part of v0.2.1 (#1812). [\[github.com\]](https://github.com/github/spec-kit/blob/main/newsletters/2026-February.md) [\[github.com\]](https://github.com/github/spec-kit/releases)
+
+**v0.3.0** (mid-March) delivered one of the most anticipated features: a **pluggable preset system** with catalog, resolver, and skills propagation (#1787). Presets let teams override Spec Kit's default templates and commands with their own conventions — a mechanism that Thulasi Rajasekaran described on LinkedIn as "the layer that makes AI-assisted development governable". The system supports priority-based stacking: an organization can layer an enterprise-standards preset beneath a team-style preset, with lower priority numbers winning conflicts. Version 0.3.0 also added a **/selftest.extension** core extension for testing other extensions against the framework (#1758), **RFC-aligned catalog integration** quality-of-life improvements (#1776), and hardened bash scripts against shell injection (#1809). On the agent side, v0.3.0 added **Mistral Vibe CLI** (#1725), migrated **Qwen Code CLI** from TOML to Markdown format (#1589/#1730), and deprecated explicit command support for the Antigravity (agy) agent (#1798/#1808). Several new community extensions arrived, including **DocGuard CDD** (#1838), **Archive & Reconcile** (#1844), **specify-status** (#1837), and **specify-doctor** (#1828). [\[github.com\]](https://github.com/github/spec-kit/releases) [\[linkedin.com\]](https://www.linkedin.com/pulse/feature-turns-spec-kit-platform-extensions-presets-rajasekaran-3ejgc)
+
+Patches came rapidly. **v0.3.1** wired **before/after hook events** into the specify and plan templates (#1886), added **JSONC deep-merge** support for `settings.json` (#1874), added the **Trae IDE** agent (#1817), and introduced priority-based resolution for both extensions and presets (#1855). A greenfield **Spring Boot pirate-speak preset** demo was published in the README (#1878), and a **Go/React brownfield walkthrough** using GitHub Copilot CLI was added to community walkthroughs (#1868). **v0.3.2** added four more new agent integrations — **JetBrains Junie** (#1831), **iFlow CLI** (#1875), and **Pi Coding Agent** (#1853) — plus migrated Codex/agy init to a native skills workflow (#1906). It also shipped a **preset submission template** (#1910) and an **Extension Comparison Guide** (#1897) to help the growing community navigate overlapping extensions. Additional community extensions added in this cycle included **verify-tasks** (#1871), **conduct** (#1908), **cognitive-squad** (#1870, updated to Triadic Model), **speckit-utils** (#1896), **spec-kit-iterate** (#1887), and **spec-kit-learn** (#1883). DocGuard received three version updates in quick succession (v0.9.8, v0.9.10, v0.9.11). [\[github.com\]](https://github.com/github/spec-kit/releases)
+
+**v0.4.0** (late March) introduced the month's headline usability feature: **auto-registration of extension skills** (#1840), so that any installed extension's commands are automatically exposed as agent skills without extra configuration. It also delivered **air-gapped/offline deployment** by embedding the core template pack directly in the CLI wheel (#1803), enabling Spec Kit to function in restricted environments with no internet access. A **timestamp-based branch naming** option was added for `specify init` (#1911) to better support parallel feature development. The YAML I/O layer was fixed to use `allow_unicode=True` and `encoding="utf-8"` (#1936), and the stale-issue GitHub Action was increased to 250 operations per run (#1922). New community extensions in this cycle included **Checkpoint** (#1947). [\[github.com\]](https://github.com/github/spec-kit/releases)
+
+Three rapid patches closed the month. **v0.4.1** fixed a missing **Assumptions section** in the spec template (#1939) and prioritized the `.specify` directory over the parent git root for **repo root detection** (#1933). **v0.4.2** was the month's most documentation-heavy release: it added **AIDE, Extensify, and Presetify** to the community catalog (#1961), moved the **community extensions table into the main README** for discoverability (#1959), added a **community presets** section (#1960), consolidated **Community Friends** sections (#1958), and formally recognized the **Spec Kit Assistant VS Code extension** (#1944). It also shipped a manual testing guide for slash command validation (#1955) and renamed "NFR" references to "success criteria" in the analyze and clarify commands (#1935). **v0.4.3** wrapped up the month by unifying Kimi/Codex skill naming and migrating legacy dotted directory names (#1971), and replacing the null-conditional operator in PowerShell scripts to restore **PowerShell 5.1 compatibility** (#1975). [\[github.com\]](https://github.com/github/spec-kit/releases)
+
+### Bug Fixes and Security Hardening
+
+Security and stability received substantial attention. The most significant fix was **shell injection hardening** of bash scripts (#1809), which addressed a class of potential injection vulnerabilities where unsanitized values from git branch names or environment variables could be passed through to shell commands. The v0.3.1 release followed up with additional bash escape and compatibility improvements (#1869). **Branch numbering** was overhauled: the old per-short-name detection scheme caused conflicts, so Spec Kit switched to **global branch numbering** (#1757) for consistent sequencing across feature branches. A **quiet git checkout** fix suppressed exceptions during branching (#1792), and a **git fetch stdout leak** was suppressed in multi-remote environments (#1876). **JSON control characters** were encoded as `\uXXXX` instead of being silently stripped (#1872). Explicit **PowerShell positional binding** was added to `create-new-feature` parameters (#1885), and the Codex native skills fallback was refreshed with legacy prompt suppression (#1930). [\[github.com\]](https://github.com/github/spec-kit/releases)
+
+### The Extension Catalog at Scale
+
+By late March, over **20 community extensions** had been built for Spec Kit. Thulasi Rajasekaran's LinkedIn article on March 20, titled *"The Feature That Turns Spec Kit Into a Platform: Extensions & Presets,"* provided a thorough analysis of what makes this ecosystem significant. Rajasekaran highlighted several standout extensions: **Conduct**, which orchestrates SDD phases by delegating to sub-agents to solve "context pollution" (where a single agent accumulates so many tokens that quality degrades); **Verify Tasks**, which scans task lists for "phantom completions" — tasks marked done with no real code behind them; **Understanding**, which runs 31 deterministic quality metrics against specifications based on IEEE/ISO standards ("like a linter for English"); and the **Jira and Azure DevOps integrations**, which auto-create work items from specs and tasks. [\[linkedin.com\]](https://www.linkedin.com/pulse/feature-turns-spec-kit-platform-extensions-presets-rajasekaran-3ejgc)
+
+Rajasekaran argued that the real significance of presets is not the mechanism itself — which is "almost comically simple" (a stack of Markdown files with a priority order) — but what it enables: the same machinery that turned "User Stories" into "Crew Tales" in the pirate-speak demo could turn them into compliance requirements with traceability IDs, add mandatory threat-model sections to every plan, or enforce TDD by requiring test tasks before implementation tasks. The pirate-speak preset, which Rajasekaran described in detail, was built by the Spec Kit maintainer using Spring Boot 4 and produced a fully functional application where every artifact — headings, paragraphs, status updates — was rendered in pirate prose. Organizations can curate which extensions are available to developers by hosting custom catalog URLs. [\[linkedin.com\]](https://www.linkedin.com/pulse/feature-turns-spec-kit-platform-extensions-presets-rajasekaran-3ejgc)
+
+## Community & Content
+
+### Developer Walkthroughs and Blog Posts
+
+March produced a wave of independent content as developers explored SDD in practice.
+
+**Tiago Valverde** published *"Spec-Driven Development in Practice: A Walkthrough with Spec Kit"* on March 14 (12-minute read). The post documents building an Instagram-style photo mural feature on a personal blog using the full Spec Kit workflow — from constitution setup through specification, clarification, planning, task breakdown, and implementation. Valverde contrasts the structured approach with previous ad-hoc prompting: while directly prompting Claude worked for small changes, for anything complex Valverde *"kept running into the same problems: scope creep mid-session, ambiguous requirements only noticed after implementation, and no artifact left behind to explain why a decision was made"*. The walkthrough includes actual Claude command syntax, expected file structures, and practical advice. Valverde recommends being *"overly specific in the initial prompt"* and to *"immediately review `spec.md`"* and add missing requirements before proceeding to planning. The clarify step is highlighted as particularly valuable — it surfaces guided questions with recommended options and identifies *"decisions that materially change complexity"*. On the broader philosophy, Valverde frames SDD as essential for production work: *"engineers still need to guarantee code quality, meet security requirements, keep documentation up to date, and keep all the stakeholders in sync with the project status and health"*. Valverde also published a shorter companion piece on March 8 titled *"The Shift from Vibe Coding to Spec-Driven Development,"* describing the broader industry trend. [\[tiagovalverde.com\]](https://www.tiagovalverde.com/posts/spec-driven-development-in-practice-a-walkthrough-with-spec-kit)
+
+**Alfredo Perez** published *"Build Your Own SDD Workflow"* on March 21, taking a deliberately contrarian approach to Spec Kit's thoroughness. Perez praises spec-driven development in principle (*"planning before you code catches bad assumptions early, keeps scope honest, and gives the AI enough context to write code you'd actually ship"*) but argues the standard seven-step workflow carries too much ceremony for smaller tasks: *"When you're adding a component or fixing a bug, that's a lot of overhead for a Tuesday afternoon"*. Perez's solution is a lean **4-step custom workflow** — `specify → plan → tasks → implement` — dropping constitution, clarify, and review. The rationale: coding rules belong in `CLAUDE.md` once (not re-checked per feature), clarification can happen iteratively rather than as a formal upfront step, and review can be added back when warranted. The custom workflow is wired into the **SpecKit Companion** VS Code extension via a `speckit.customWorkflows` configuration in `.vscode/settings.json`, giving users a visual sidebar where each phase appears as a clickable step. Perez walks through a real feature implementation — adding a home page and navigation bar — showing each phase in action with screenshots of the SpecKit Companion panel. The article highlights an important tradeoff: **full rigor vs. lightweight adoption**. Not every project needs all seven steps, and Spec Kit's extensible design accommodates both extremes. Perez also presented this workflow live at an **Angular Community Meetup** on March 25 titled *"Create Your Own AI Workflow"*. [\[alfredo-perez.dev\]](https://www.alfredo-perez.dev/blog/2026-03-21-build-your-own-sdd-workflow)
+
+**Sergey Golubev** of prodfeat.ai published *"20+ SDD Frameworks: A Catalog for AI Development"* on March 17, the result of two weeks of research across YouTube, Telegram, Medium, Augment Code, ThoughtWorks Radar, and GitHub repositories. The catalog organizes **20+ frameworks in 6 categories** and highlights three standouts. **BMAD-METHOD** (\~41,000 stars) simulates an entire agile team from AI roles — Analyst, PM, Architect, Scrum Master — and produces a full PRD with architecture and dev stories; Golubev calls it *"perfect for a team of 3-5 people"* but notes a high barrier to entry. **QuintCode + FPF** is notable for its ADI Cycle in 5 phases (hypothesize, verify logic, test, audit, decide), which preserves decision rationale — *"three months later you won't remember the reasoning."* In one case study, FPF + ChatGPT Pro produced a 52-page spec and 280 feature files in two evenings, and QuintCode chose Docker Swarm over Kubernetes through reasoned analysis. **cc-sdd** (\~2,880 stars) provides Kiro-style SDD commands for 8 tools (Claude Code, Cursor, Gemini CLI, Codex CLI, and 4 more), with an enforced workflow that *"won't let you skip planning"*. [\[prodfeat.ai\]](https://www.prodfeat.ai/en/blog/2026-03-17-sdd-frameworks-catalog)
+
+Golubev also presents a **three-level SDD maturity model**: *Spec-First* (spec per task, discarded after implementation), *Spec-Anchored* (spec as a living document, changes start from it), and *Spec-as-Source* (spec is the only artifact, code is compiled output). At the most aggressive end, **Tessl** ($125M raised) is building toward spec-as-source: *"code becomes a byproduct of specification"*. Golubev's conclusion: *"SDD is not a fad and not waterfall in markdown… AI agents generate good code when the task is well-defined. Without a spec — you're rolling the dice"*. The article references the **METR study (July 2025)** finding that developers using AI were **19% slower** on real-world tasks, attributing the problem to *"debugging loops from unstructured prompts"* — a finding that has become a recurring justification for the SDD movement. [\[prodfeat.ai\]](https://www.prodfeat.ai/en/blog/2026-03-17-sdd-frameworks-catalog)
+
+### Community Tools and Documentation Improvements
+
+The **Spec Kit Assistant VS Code extension**, a third-party tool providing a graphical interface for running Spec Kit commands, was formally recognized as a "Community Friend" and added to the project README (#1944, #1956). The README underwent significant reorganization during the month: the team consolidated **Community Friends** sections (#1958), moved the **community extensions table** into the main README for discoverability (#1959), added a **community presets** section (#1960), an **AIDE extension demo** (#1943), and updated the publishing guide with Category and Effect columns to help extension authors categorize submissions (#1913). An **Extension Comparison Guide** was also published (#1897) and a manual testing guide for slash command validation (#1955). The team added multiple technology-specific walkthroughs: a **Java brownfield walkthrough** (#1820), a **Go/React brownfield dashboard walkthrough** (#1868), and the **Spring Boot pirate-speak preset** demo (#1878), supplementing the walkthroughs for .NET CLI, Spring Boot + React, and ASP.NET CMS that were committed in late February/early March. [\[github.com\]](https://github.com/github/spec-kit/releases)
+
+A notable community project appeared on GitHub: **speckit-pipeline** by iandeherdt, described as *"a pipeline on top of spec-kit to automate the design and build process with an evaluation step"*. This tool installs specialized Claude Code agents for a **design loop** (designer + design-critic agents iterating in a real browser) and a **build loop** (developer + evaluator agents verifying implementations against acceptance criteria). Both loops produce structured feedback and iterate up to 5 cycles per sprint until the work passes a quality gate. While small (4 commits, 0 stars as of retrieval), it represents the kind of higher-order automation that the community is building atop Spec Kit's foundation. The official Spec Kit repository has an open issue (#1966) requesting a built-in pipeline command for automated end-to-end workflow execution, suggesting this pattern may eventually be incorporated into the core. [\[github.com\]](https://github.com/iandeherdt/speckit-pipeline)
+
+A public **Microsoft Learn** training module titled *"Implement Spec-Driven Development using the GitHub Spec Kit"* was also available during March — a 3-hour, 13-unit, intermediate-level course covering practical SDD workflows with Spec Kit, providing an onboarding path for enterprise developers encountering SDD for the first time.
+
+## SDD Ecosystem & Industry Trends
+
+### The "Vibe Coding Is Dead" Narrative
+
+On March 20, *ByteIota* published *"Spec-Driven Development Kills 'Vibe Coding'"*, reporting that AWS developers were pushing SDD as the new standard for AI-assisted coding. The article cited **over 100,000 developers** adopting SDD approaches in the first five days of tool previews, AWS demonstrating a two-week feature completed in **two days** using Kiro IDE with structured specs, and World Economic Forum research indicating **65% of developers** expect their role to change around spec-first workflows in 2026. Spec Kit received a direct recommendation as a free, open-source option supporting **22+ AI platforms**. [\[byteiota.com\]](https://byteiota.com/spec-driven-development-kills-vibe-coding-march-2026/)
+
+The article gave equal space to critics. *Marmelab* called SDD *"the exact mistakes Agile was designed to solve."* A controlled test by *Isoform* found SDD took **33 minutes** to produce **689 lines of code** versus **8 minutes with iterative prompting**, with no measured quality improvement. The emerging consensus favored **hybrid approaches** — a Red Hat developer recommendation captured the middle ground: *"Use the vibes to explore. Use specifications to build."* Reported success cases included Google's migrations (**50% time reduction**), Airbnb migrating **3,500 test files** at a **15× speedup**, and API-first teams achieving **75% cycle time reduction**. But solo developers consistently reported overhead exceeding benefit. [\[byteiota.com\]](https://byteiota.com/spec-driven-development-kills-vibe-coding-march-2026/)
+
+The ByteIota piece was not isolated. Within a two-week window, independent articles appeared from **Shimon Ifrah** (*"Vibe Coding Is Dead"*), **Raul Proenza** at Cox Automotive (*"Welcome to Agentic Engineering"*), **CGI** (*"From vibe coding to intent engineering"*), and **Vishal Mysore** on Medium (*"A Map of 30+ Agentic Coding Frameworks"*). ByteIota also raised an underappreciated concern: if specifications replace coding as the primary activity, **how do junior developers build the judgment needed to write good specs or review AI-generated code?** No proven onboarding model exists yet for a spec-first world. [\[byteiota.com\]](https://byteiota.com/spec-driven-development-kills-vibe-coding-march-2026/)
+
+### Competitive Landscape and Comparisons
+
+The SDD tool ecosystem expanded rapidly. Beyond the frameworks catalogued by Golubev, March saw deeper public analysis of how tools compare.
+
+On March 31, **Augment Code** published *"Intent vs GitHub Spec Kit (2026): Platform or Framework?"* The analysis framed Spec Kit as an agent-agnostic framework producing **plain Markdown files identical across agents**, while Intent (Augment Code's own product) offers **"living specs"** that auto-update as agents complete work. The core tradeoff: Spec Kit's strength is **portability** across 22+ agents; Intent's is **deeper native integration** with automated drift detection. The comparison surfaced the **drift problem** as a key architectural concern — Spec Kit's specs can become stale post-implementation, and while community extensions (Retrospective, Verify, Sync) address this post-facto, native real-time drift detection is not yet in core. [\[augmentcode.com\]](https://www.augmentcode.com/tools/intent-vs-github)
+
+The broader competitive landscape continued to evolve. **OpenSpec** (Fission AI) remained at ~29,300 stars, **BMAD-METHOD** grew to ~41,000 stars, and **Tessl** continued in private beta pursuing spec-as-source. AWS's **Kiro** delivered the two-day implementation result cited by ByteIota. While Spec Kit leads in GitHub popularity and agent breadth, alternatives are differentiating on orchestration depth (Intent, BMAD), enforced planning discipline (cc-sdd), decision audit trails (QuintCode), and spec-as-source vision (Tessl). [\[prodfeat.ai\]](https://www.prodfeat.ai/en/blog/2026-03-17-sdd-frameworks-catalog)
+
+## Roadmap
+
+Areas under discussion or in progress for future development:
+
+- **Spec lifecycle management** -- supporting longer-lived specifications that evolve across multiple iterations. The Augment Code comparison and community commentary highlighted "spec drift" as a key concern. The Archive & Reconcile extension (#1844) is a community step; a core solution is expected to be a focus area. [\[augmentcode.com\]](https://www.augmentcode.com/tools/intent-vs-github) [\[github.com\]](https://github.com/github/spec-kit/releases)
+- **CI/CD integration** -- incorporating Spec Kit verification into pull request workflows and failing builds when specs are out of alignment. The Jira and Azure DevOps extensions (#1764, #1734) are a first step. [\[github.com\]](https://github.com/github/spec-kit/releases)
+- **End-to-end workflow automation** -- an open issue (#1966) proposes a built-in pipeline command. The community-built **speckit-pipeline** by iandeherdt already demonstrates multi-agent loops with browser verification. [\[github.com\]](https://github.com/iandeherdt/speckit-pipeline)
+- **Continued agent expansion** -- seven new agents were added in March alone. The agent-agnostic design means support for emerging tools can be added by anyone. [\[byteiota.com\]](https://byteiota.com/spec-driven-development-kills-vibe-coding-march-2026/)
+- **Experience simplification** -- the preset system, custom workflows, and growing walkthrough library lower the learning curve, but extension discoverability will need a more robust solution as the catalog grows. [\[github.com\]](https://github.com/github/spec-kit/releases)
+- **Toward a stable release** -- nine releases in one month reflects pre-1.0 momentum. Reaching 1.0 will require stabilizing the extension and preset APIs and ensuring backward compatibility across the agent and extension surface area. [\[github.com\]](https://github.com/github/spec-kit/blob/main/newsletters/2026-February.md)
+
+