# getunblocked.com

> AI-optimized mirror of getunblocked.com containing 50 pages totalling 61,992 words of clean markdown content, structured data, and semantic HTML. Original source: https://getunblocked.com. Last updated: 2026-08-09T03:03:57.512Z.

Each page is available as HTML (with JSON-LD structured data) and Markdown (text-only, ideal for LLMs and RAG).

## Homepage

- [The context engine for modern engineering teams | Unblocked](/content/site-root.html): Unblocked turns code, docs, tickets, and conversations into actionable context — engineers move faster and AI agents generate mergeable code on the first try. (633 words)

## Articles & Blog Posts

- [Context Engineering: Why LLMs need more than prompts and MCP servers](/content/blog/context-engineering/index.html): Context engineering is shaping what an AI agent sees before it acts — the principles, the maturity model, and why prompts and MCP alone fall short. (4,648 words)
- [AI Coding Tool ROI: When Finance Audits the Spend (2026)](/content/blog/ai-coding-tool-roi-reckoning/index.html): AI coding tool spend is exploding as 95% of GenAI pilots show no P&L impact. The 2026 ROI audit playbook for engineering leaders before finance asks. (2,567 words)
- [AI Agent Doom Loop: Why Agents Repeat the Same Mistake](/content/blog/ai-agent-doom-loop/index.html): 82% of developers hit repeated AI agent failures. Learn why the AI agent doom loop happens and what breaks the cycle of agents making the same mistakes. (2,923 words)
- [How to Measure AI Coding Tool ROI (A VP's Formula)](/content/blog/measure-ai-coding-tool-roi/index.html): Atlassian found 99% of devs save time with AI, but 50% lose 10+ hours a week. Here is a five-step formula to measure AI coding tool ROI honestly. (2,450 words)
- [AI Tokenomics: Measure What AI Tools Cost](/content/blog/ai-tokenomics-cost-framework/index.html): AI tokenomics explained: enterprise LLM spend doubled to $8.4B in six months while only 39% of orgs see EBIT impact. Measure token yield, not just price. (2,637 words)
- [AI Legacy Modernization: The Complete 2026 Guide](/content/blog/legacy-code-modernization/index.html): Legacy modernization with AI works or fails on context: Meta took agent-usable coverage from 5% to 100%. A 2026 playbook for engineering leaders. (2,739 words)
- [The Auto-Loop Tax: AI Agent Token Cost, 15x](/content/blog/agent-auto-loop-token-cost/index.html): Self-running AI agents burn roughly 15x the tokens of a chat. Why unbounded loops raise AI agent token cost without raising yield, and the discipline that fixes it. (2,434 words)
- [MCP Tool Overload: A Measured Guide to Context-Window Bloat](/content/blog/mcp-tool-overload/index.html) (2,473 words)
- [The Context Problem Moved From People to Agents](/content/blog/the-context-problem-moved-from-people-to-agents/index.html): 82% of developers use AI tools weekly but only 33% trust accuracy (Stack Overflow 2025). The context problem shifted from people to agents. (2,219 words)
- [Three Hard Lessons from Building Context at Scale (2026)](/content/blog/three-hard-lessons-context-scale/index.html): We optimized for access, hid conflicts, and cached answers. Three retrospective lessons from building a production context engine for AI coding agents in 2026. (1,791 words)
- [What is a context engine? Why ‘context-aware’ dev tools fall short](/content/blog/context-aware-dev-tools-arent-quite-there-yet/index.html): ‘Context-aware’ AI dev tools mostly add access to more data, not understanding. A context engine turns scattered information into decision-grade context. (1,347 words)
- [Building a Context Engine on Top of MCP (2026 Guide)](/content/blog/building-a-context-engine-on-mcp/index.html): MCP is a transport, not a context engine. 3 patterns to wire reasoning above MCP, 4 spec gaps, and a build-vs-buy rubric for engineering leaders in 2026. (1,997 words)
- [ServiceNow vs Tabnine vs Unblocked: Context Engines](/content/blog/servicenow-vs-tabnine-vs-unblocked/index.html): 76% of devs use AI tools, but few orgs govern context. Compare ServiceNow, Tabnine, and Unblocked context engines across 9 enterprise dimensions. (2,024 words)
- [Refactoring Legacy Code with AI Agents: Safe Workflow](/content/blog/refactoring-legacy-code/index.html): A 2025 study found AI made legacy refactoring faster but no smarter. Learn the 4-step workflow to refactor legacy code with AI agents safely. (1,848 words)
- [Context Engineering Grew Up at the World's Fair](/content/blog/context-grew-up-at-the-worlds-fair/index.html): At the 2026 AI Engineer World's Fair, two frontier model launches drew real excitement, yet 6,000 engineers kept circling back to context. Why context is now the bottleneck. (1,361 words)
- [The Outer Loop Needs a Context Engine, Not More Reviewers](/content/blog/the-outer-loop-needs-a-context-engine/index.html): AI teams merge 98% more PRs but review time is up 91%. The outer loop is the new bottleneck — here's why more review and bigger context windows don't fix it. (1,695 words)
- [The MCP Plateau: Why More MCP Servers Don't Fix Agent Output](/content/blog/mcp-plateau/index.html): Teams wire up 10+ MCP servers and output quality flatlines. Anthropic measured 150,000 tokens spent before work began. The 6 things a context engine does that MCP can't. (1,754 words)
- [Why a 12-Line Pull Request Can Take 5 Days (It’s a Context Problem)](/content/blog/developer-productivity-is-still-broken/index.html): A 12-line pull request took 5 days — not to write, but to gather context. Why small code changes take days, and why AI code tools don’t fix it. (1,286 words)
- [How to give your AI coding agent your company’s context](/content/blog/what-your-coding-agent-cant-see/index.html): AI coding tools read your code but not the decisions behind it. Here’s how to give your coding agent your company’s context, via Slack, PRs, and tickets. (2,127 words)
- [“I haven’t found a tool that does context as effectively as Unblocked does.”](/content/customers/advidi/index.html): Advidi connected its engineers and its homegrown AI agents, like the OpsClaw incident bot, to one context engine spanning 400+ repositories, Slack, Confluence, and Jira, through a single MCP. (1,340 words)
- [Stale data isn't stale: agent retrieval and pruning context](/content/blog/stale-data-isnt-stale/index.html): Across 17,190 user-rated references, helpfulness rate stays flat from same-day sources back through 10-year-old ones. Don't prune by age. (1,463 words)
- [Unblocked vs Sourcegraph Cody: Code Context for Enterprises](/content/blog/unblocked-vs-sourcegraph-cody/index.html): Developers spend 60% of their time reading code. Compare Sourcegraph Cody and Unblocked on context depth, data sources, and enterprise fit. (1,616 words)
- [Why MCP isn’t enough for coding agents — and what fixes it](/content/blog/why-mcp-isnt-enough-enhancing-agent-capabilities-with-a-context-engine.html): MCP gives agents access to more sources but not the judgment to use them. Here’s why MCP isn’t enough on its own, and how a context engine helps. (1,704 words)
- [Your new developer feels like a fraud. It's your onboarding process.](/content/blog/no-imposter-onboarding/index.html): New engineers face a tough learning curve, navigating a new role and a complex codebase. It’s no wonder they feel like imposters. We need to reimagine onboarding to better support them. (1,670 words)
- [Agent Memory Hub Explained: TencentDB-Agent-Memory](/content/blog/team-memory-hubs-ai-agents/index.html): TencentDB-Agent-Memory passed 12,800 GitHub stars. What an agent memory hub solves for teams, what it can't adjudicate, and when you need a context engine. (1,408 words)
- [Introducing new features that make Unblocked even more helpful in Slack](/content/blog/slack-dms-and-private-channels/index.html): We’re expanding the ways teams can get fast, accurate, and relevant answers to their questions in Slack with auto responses, support for private channels, bot DMs, and channel summarizations. (924 words)
- [Asana MCP integration for AI agents | Unblocked](/content/integrations/asana/index.html): Connect Asana to Unblocked so AI coding agents get cited context from your projects, tasks, and comments alongside your code. (684 words)
- [Slack MCP integration for AI agents | Unblocked](/content/integrations/slack/index.html): Connect Slack to Unblocked so AI coding agents get cited context from your channels and threads, and your team can ask Unblocked questions in Slack. (698 words)
- [Cribl: “Unblocked is our number one tool for finding the information we should know but don’t.”](/content/customers/cribl/index.html): Cribl’s remote-first culture empowers its employees with flexibility, but keeping over 100 engineers on the same page across time zones poses a real challenge. That’s why the team turned to Unblocked. (921 words)
- [OSS Repo Rules Agent: give your coding agent a memory of your team's rules](/content/blog/oss-repo-rules-agent/index.html): We're open-sourcing repo-rules-agent, a CLI tool that turns CLAUDE.md, AGENTS.md, .cursorrules, and ~40 other rules files into a queryable index any coding agent can consult on demand. (716 words)
- [Google Drive MCP integration for AI agents | Unblocked](/content/integrations/google-drive/index.html): Connect Google Drive to Claude Code, Cursor, Copilot, and other AI agents. Search selected Docs, Sheets, Slides, text files, and PDFs with source citations. (781 words)
- [Bitbucket MCP integration for AI agents | Unblocked](/content/integrations/bitbucket/index.html): Connect Bitbucket to Unblocked to give AI coding agents cited context from your repositories, pull requests, and code history. (727 words)
- [Reduce rework. Ship faster | Unblocked context engine](/content/outcomes/reduce-rework/index.html): Agents with decision-grade context get it right the first time instead of looping on corrections — so you spend less time reviewing and more time shipping. (303 words)
- [Onboarding new developers with Unblocked](/content/use-cases/onboarding/index.html): Unblocked helps new team members find answers they need so they can start contributing right away. (491 words)
- [Jira MCP integration for AI agents | Unblocked](/content/integrations/jira/index.html): Connect Jira to Unblocked so AI coding agents get cited context from your projects and issues, connected to the code and PRs behind them. (711 words)
- [Mergeable by default: Building the context engine to save time and tokens — Unblocked](/content/videos/mergeable-by-default-building-the-context-engine-to-save-time-and-tokens.html): Agents can generate code. The hard part is generating code that's right for your system, team conventions, and past decisions. That's a context problem that naive RAG, MCP servers, and bigger context windows don't solve. Without the right context, that code costs you twice: once in tokens, again in long review cycles. This session is a practitioner's guide to building a context engine: the reasoning layer that brings together your organizational context and delivers only what the agent needs for the task at hand. Peter walks through the challenges that matter: reasoning across conflicting sources, maintaining permissions, and personalizing results based on who's asking and what they're working on, going deep on specific components with live demos and technical breakdowns. Drawn from real lessons building this in production, including what went wrong. (162 words)
- [How to stop babysitting your agents — Unblocked](/content/videos/how-to-stop-babysitting-your-agents/index.html): Your agents are fast, capable, and completely context-blind. They generate code that compiles but doesn't reflect how your system actually works. You're likely already feeling it: ballooning token costs, longer review cycles, inconsistent output. And as the industry shifts toward autonomous background and cloud agents, these problems only compound. More MCPs, more rules, and bigger context windows give agents access to information — not understanding. The teams pulling ahead have something different: a context engine that gives agents exactly what they need for the task at hand. What we'll cover: Where teams get stuck on the AI maturity curve, and why progress stalls Why naive RAG, bigger context windows, and more rules don't solve the root problem What a context engine actually is, and how it works in practice Real-world lessons from building a context engine at scale A live demo: the same coding task, with and without organizational context You'll walk away with: A clear mental model for making AI agents context-aware Practical ways to cut token costs and review cycles A framework for improving output quality without overloading context A look at how leading teams are scaling AI effectively Ideas you can put to work in your own workflows this week (232 words)
- [Security & Compliance: SOC 2, CASA, GDPR — Unblocked](/content/security/index.html): Unblocked keeps your data isolated, authorized, and encrypted, and never used to train shared models. SOC 2 Type II, CASA Tier II, and GDPR compliant. (435 words)
- [Building the Context Engine AI Agents Need (AI Dev 26 x SF) — Unblocked](/content/videos/building-the-context-engine-ai-agents-need-ai-dev-26-x-sf.html): Every AI coding tool can generate code. Very few can generate the right code for your organization, because they're missing context. They don't know why your team chose Redis over DynamoDB, what the team decided in a Slack thread from two months ago about the auth migration, or which architectural patterns your principal engineers actually enforce in review. Brandon Waselnuk from Unblocked shares a practitioner's guide to building a context engine: the reasoning layer that continuously synthesizes organizational knowledge across disparate sources into unified, queryable understanding. Brandon walked through the problems you actually have to solve — reasoning across systems that don't agree with each other, searching globally before you can reason, maintaining identity-scoped permissions so every user and agent only sees what they should, and personalizing results based on who's asking and what they're working on. These are the engineering challenges that make naive RAG fall short, drawn from real lessons building this at scale. (188 words)
- [Stop babysitting your agents... — Unblocked](/content/videos/stop-babysitting-your-agents/index.html): Same prompt. Same agent. Same model. Without a context engine: 2.5 hours, 20.9 million tokens, multiple rounds of human correction, and code that compiled but would have broken the entire system if it shipped. With one: 25 minutes, 10.8 million tokens, and a senior engineer who gave one nitpick and approved the merge. Brandon Waselnuk from Unblocked makes the case that the problem is not access but understanding. More MCPs give agents pipes to information. A million token context window just sits there. Naive RAG stops at the first result it finds, a phenomenon called satisfaction of search borrowed from radiology. What actually changes is a context engine that reasons across your codebase, Slack history, PR patterns, and org structure to build a research packet before the agent starts writing, so it arrives knowing your factory patterns, your fallback infrastructure, and what the CTO said was wrong in that thread three months ago. (172 words)
- [AI coding agents are always the new hire — Unblocked](/content/videos/ai-coding-agents-are-always-the-new-hire/index.html): In this episode, Dennis Pilarinos, founder and CEO of Unblocked, joins us at LeadDev in London. Dennis explains why AI coding agents are only as good as the context they can access, and why every new agent is like a developer on their first day at a company. We talk about organizational context, token savings, MCP, stale documentation, knowledge graphs, permissions, identity across systems, and why the next big infrastructure layer for AI development may be the context layer. (104 words)
- [Buildkite CI failure analysis | Unblocked](/content/integrations/buildkite/index.html): Connect Buildkite to Unblocked for context-aware CI failure analysis and root-cause reports in pull requests and merge requests. (525 words)
- [Building the context engine AI agents need — Unblocked](/content/videos/building-the-context-engine-ai-agents-need/index.html): Every AI coding tool can generate code. Very few can generate the right code for your organization, because they're missing context. This talk is a practitioner's guide to building a context engine: the reasoning layer that continuously synthesizes organizational knowledge across disparate sources into unified, queryable understanding. (71 words)
- [Cut Your AI Agents' Token Costs by 30% | Unblocked MCP in Action — Unblocked](/content/videos/cut-your-ai-agents-token-costs-by-30-unblocked-mcp-in-action.html): Every new coding agent starts without the organizational context your engineers already have. It spends time searching repositories, documentation, and tickets before it can begin implementing. That discovery burns both time and tokens. In this 85-second walkthrough, Brandon shows how feeding your agents organizational context through Unblocked lets them skip the search phase and reduce token costs by up to 30%. (80 words)
- [Exploring context in AI and devtools with Unblocked founding engineers — Unblocked](/content/videos/exploring-context-in-ai-and-devtools-with-unblocked-founding-engineers.html): Join the Unblocked team as they dive deep into the concept of 'context' and its critical role in AI and developer tooling. Founding engineers Peter, David, and Richie discuss their unique challenges and solutions in developing a powerful context engine at Unblocked. Discover how they handle integration, access control, and focus to enhance the efficiency of engineering processes. Learn about the importance of curating relevant information and minimizing noise to ensure optimal performance in AI applications and agentic coding tools. Perfect for developers looking to understand the intricacies of AI-powered tools. Read the companion blog post: What happens when you put three AI tool builders in a room together (134 words)
- [Building the Context Engine AI Agents Need — Unblocked](/content/videos/building-the-context-engine-ai-agents-need-infoq/index.html): AI coding agents aren't failing because they lack intelligence — they're failing because they lack context. Dennis Pilarinos introduces the context engine: a reasoning layer that continuously synthesizes organizational knowledge, resolves conflicting sources, and delivers identity-scoped, token-optimized context so agents produce correct, org-aware code instead of triggering costly correction loops. (68 words)
- [Context Engine | Unblocked](/content/context-engine/index.html): Turn scattered data into curated context. Unblocked's Context Engine connects your codebase, docs, and conversations into a unified knowledge graph for coding agents. (492 words)
- [Unblocked: Building a context engine for developers — Unblocked](/content/videos/unblocked-building-a-context-engine-for-developers/index.html): In this episode, Peter, a founding engineer at Unblocked, discusses his role in developing the intelligence pipeline for the platform. He digs into his history of technology startups, including Buddy Build (Acquired by Apple), and their journey to building Unblocked—a context engine designed to surface all necessary context for developers and their agents while filtering out the noise. Peter explains how Unblocked builds social and expert graphs to resolve conflicts and provide accurate context. They explore the importance of understanding user intent, managing permissions, and dealing with outdated documentation. As they dig deeper, Peter touches on advanced topics like conflict resolution, incident management, and the future of AI in software development. This conversation provides valuable insights for developers interested in optimizing their workflows and increasing efficiency in their engineering tasks and working with agentic coding tools. Read the companion blog post: How a context engine actually works, and why you need to care now (176 words)
- [The Unblocked Blog](/content/blog/category-ai-code-review/index.html): Insights on AI coding agents, developer productivity, and engineering team context from the Unblocked team. (43 words)

## Resources

- [Full Page Index](/index.html): Browse all cached pages with rich metadata
- [About This Cache](/about.html): Methodology, technical details, and usage guidelines
- [XML Sitemap](/sitemap.xml): Machine-readable sitemap for crawler discovery
- [Robots.txt](/robots.txt): Crawler directives