# Agent Primer — Matt Pocock on Agentic Engineering, Custom Skills, and Sandcastle > **Read me first.** This document primes a downstream AI agent to act as a subject-matter expert on the source video. Read this in full before consulting individual notes. **Source**: [Matt Pocock on Agentic Engineering, Custom Skills, and Sandcastle](https://www.youtube.com/watch?v=nQwJVHCtDDY) **Duration**: 51m 40s **Speakers**: Matt Pocock, David (Interviewer) **Domains**: `ai-coding-agents`, `software-architecture`, `developer-productivity`, `agentic-workflows`, `developer-tools` **Vault slug**: `matt-pocock-agentic-engineering-sandcastle` **Generated**: 2026-06-23T07:49:02.924Z --- # Agent Primer — Matt Pocock on Agentic Engineering, Custom Skills, and Sandcastle > This document is your single most important briefing. After reading it you should be able to answer ~80% of questions about the source video without consulting other notes. Cross-references use [[wikilink-id]] syntax. ## 1. Source at a glance - **Video:** *Matt Pocock on Agentic Engineering, Custom Skills, and Sandcastle* - **Duration:** ~52 minutes (3,100 seconds) - **Speakers:** - [Matt Pocock](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/entities/entity-matt-pocock.md) — TypeScript educator (Total TypeScript), former Stately/Vercel, creator of [Sandcastle](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/entities/entity-sandcastle.md) and the [`mattpocock/skills`](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/entities/entity-matt-pocock-skills.md) repo. The primary voice and the subject-matter expert. - [David](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/entities/entity-david-interviewer.md) — interviewer/host. Drives the conversation but does not advance independent claims. - **Domain tags:** ai-coding-agents, software-architecture, developer-productivity, agentic-workflows, developer-tools. ## 2. The single thesis (memorize this) > To unlock the true potential of AI coding agents, developers must shift their focus from obsessing over the latest underlying models to optimizing the **harness** — the surrounding environment, tools, and procedural skills. Because AI has effectively commoditized **tactical** programming (syntax and localized bug fixing), human developers must elevate themselves to **strategic** programmers who architect codebases specifically for AI navigability. Furthermore, true agentic workflows require moving away from synchronous chat and infinite loops toward deterministic, queue-based **AFK** (Away From Keyboard) execution within secure, isolated sandboxes. Three pillars rolled into one thesis: 1. **Harness over model.** Optimize what surrounds the LLM, not the LLM itself. 2. **Strategic over tactical.** AI ate boilerplate; humans must move up the stack. 3. **AFK + queues + sandboxes.** The future of agentic work is asynchronous, decomposed, and isolated. ## 3. Mental model for the vault The vault is organized around five tightly interwoven ideas. If you internalize this graph you understand the source: ``` [concept-ai-harness](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/concepts/concept-ai-harness.md) ←—— [claim-harness-over-model](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/claims/claim-harness-over-model.md) ←— [quote-f1-harness-analogy](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/quotes/quote-f1-harness-analogy.md) │ │ ├──> [entity-sandcastle](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/entities/entity-sandcastle.md) ────┴──> [framework-afk-agent-pipeline](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/frameworks/framework-afk-agent-pipeline.md) ├──> [entity-matt-pocock-skills](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/entities/entity-matt-pocock-skills.md) └──> [concept-procedural-vs-ability-skills](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/concepts/concept-procedural-vs-ability-skills.md) ──> [claim-procedural-over-abilities](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/claims/claim-procedural-over-abilities.md) [concept-tactical-vs-strategic-programming](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/concepts/concept-tactical-vs-strategic-programming.md) ←— [entity-a-philosophy-of-software-design](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/entities/entity-a-philosophy-of-software-design.md) / [entity-john-ousterhout](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/entities/entity-john-ousterhout.md) │ ├──> [claim-ai-eaten-tactical](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/claims/claim-ai-eaten-tactical.md) ←— [quote-ai-eaten-tactical](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/quotes/quote-ai-eaten-tactical.md) ├──> [claim-skills-are-ceiling](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/claims/claim-skills-are-ceiling.md) ←— [quote-skills-are-ceiling](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/quotes/quote-skills-are-ceiling.md) └──> [framework-strategic-ai-delegation](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/frameworks/framework-strategic-ai-delegation.md) [concept-afk-agent-work](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/concepts/concept-afk-agent-work.md) ←—— [concept-agentic-queues](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/concepts/concept-agentic-queues.md) ←— [claim-queues-over-loops](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/claims/claim-queues-over-loops.md) │ │ └──> [framework-afk-agent-pipeline](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/frameworks/framework-afk-agent-pipeline.md) └──> rejects [entity-ralph](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/entities/entity-ralph.md) / Auto-GPT-style loops [concept-stateful-learning-skills](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/concepts/concept-stateful-learning-skills.md) ── uses [concept-zone-of-proximal-development](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/concepts/concept-zone-of-proximal-development.md) ── lives in [entity-matt-pocock-skills](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/entities/entity-matt-pocock-skills.md) Counter-pressure: [entity-the-bitter-lesson](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/entities/entity-the-bitter-lesson.md) ──> [question-ai-vs-bitter-lesson](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/open-questions/question-ai-vs-bitter-lesson.md) ``` ## 4. Core concepts (with [[wikilinks]]) ### 4.1 [The AI Harness](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/concepts/concept-ai-harness.md) The environment around the LLM: tools, memory, control loop, quality gates, skills, sandboxes. Pocock's canonical metaphor (the F1 car: engine vs. chassis/aerodynamics/steering) lives in [quote-f1-harness-analogy](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/quotes/quote-f1-harness-analogy.md). Developers control the harness; they don't control the model. Therefore the harness is where ROI lives — for now. ### 4.2 [Tactical vs. Strategic Programming](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/concepts/concept-tactical-vs-strategic-programming.md) Borrowed from [Ousterhout](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/entities/entity-john-ousterhout.md)'s [A Philosophy of Software Design](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/entities/entity-a-philosophy-of-software-design.md). Tactical = syntax, bug fixes, localized changes. Strategic = architecture, interfaces, long-term velocity. Pocock claims AI has *eaten* tactical (see [claim-ai-eaten-tactical](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/claims/claim-ai-eaten-tactical.md)), so human value is now strictly strategic. ### 4.3 [Procedural vs. Ability Skills](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/concepts/concept-procedural-vs-ability-skills.md) **Abilities** = the *model* autonomously invokes a tool when it sees fit. **Procedures** = the *human* invokes a tool explicitly (typically as a slash command). Pocock strongly prefers procedures and actively *disables* model auto-invocation. Operational directive: [action-blank-slate-agents](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/action-items/action-blank-slate-agents.md). Contrarian framing: [contrarian-disable-model-skills](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/contrarian-insights/contrarian-disable-model-skills.md). ### 4.4 [AFK Agent Work](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/concepts/concept-afk-agent-work.md) Away From Keyboard. Agents work autonomously in the background, picking tasks from a queue, executing inside sandboxes, opening PRs. The full pipeline is [framework-afk-agent-pipeline](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/frameworks/framework-afk-agent-pipeline.md); the execution layer is [entity-sandcastle](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/entities/entity-sandcastle.md). ### 4.5 [Agentic Queues vs. Loops](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/concepts/concept-agentic-queues.md) Loop-based agents (Auto-GPT, BabyAGI, [entity-ralph](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/entities/entity-ralph.md)) are non-deterministic and hard to debug. Queue-based agents — one scoped task → one PR → stop — mirror how human engineering teams work. The contrarian framing is [contrarian-queues-not-loops](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/contrarian-insights/contrarian-queues-not-loops.md); the claim is [claim-queues-over-loops](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/claims/claim-queues-over-loops.md). ### 4.6 [Stateful Learning Skills](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/concepts/concept-stateful-learning-skills.md) Pocock's `teach` skill writes state to disk (`MISSION.md`, learning record, HTML cheat sheets, quizzes) so the LLM becomes a persistent tutor. Grounded in [Vygotsky's Zone of Proximal Development](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/concepts/concept-zone-of-proximal-development.md). Pocock's prior decade as a teacher (and earlier as a vocal coach) informs this design. ## 5. Top claims (with confidence levels) | # | Claim | Confidence | Testable | Note | |---|-------|-----------|----------|------| | 1 | [The harness outweighs the model](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/claims/claim-harness-over-model.md) | High | Yes | Defensible directionally; absolute form challenged by [entity-the-bitter-lesson](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/entities/entity-the-bitter-lesson.md). | | 2 | [AI has eaten tactical programming](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/claims/claim-ai-eaten-tactical.md) | High (rhetorical) | No | Directionally true; \"eaten\" is overstated per enrichment overlay. | | 3 | [Human strategic skills dictate the AI ceiling](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/claims/claim-skills-are-ceiling.md) | High | Yes | Consistent with human-AI collaboration literature; precise 10x is speculative. | | 4 | [Procedural skills beat autonomous abilities](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/claims/claim-procedural-over-abilities.md) | Medium | Yes | Context-dependent; autonomous frameworks (AutoGen, LangGraph) are also viable. | | 5 | [Queues are superior to agent loops](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/claims/claim-queues-over-loops.md) | High | Yes | Strong support from distributed-systems wisdom. | | 6 | [Enthusiasm beats experience](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/claims/claim-enthusiasm-beats-experience.md) | Medium | No | Anecdotal; qualified by a fundamentals floor. | ## 6. Frameworks ### 6.1 [Strategic AI Delegation Framework](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/frameworks/framework-strategic-ai-delegation.md) Five steps to delegate tactical work safely: 1. Design hard architectural pieces upfront. 2. Scope tasks into discrete units. 3. Define clear module interfaces. 4. Create test seams and scenes. 5. Maintain context-pointer documentation. ### 6.2 [AFK Agent PR Pipeline](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/frameworks/framework-afk-agent-pipeline.md) Six-stage production pipeline: 1. Issue queue. 2. Orchestrator spins up [Sandcastle](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/entities/entity-sandcastle.md) sandbox. 3. Agent implements. 4. Agent opens PR. 5. Secondary CI agent reviews. 6. Human merges. The human's role becomes engineering manager, not author. ## 7. Key entities - **People:** [Matt Pocock](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/entities/entity-matt-pocock.md), [David (Interviewer)](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/entities/entity-david-interviewer.md), [John Ousterhout](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/entities/entity-john-ousterhout.md). - **Tools/products:** [Sandcastle](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/entities/entity-sandcastle.md) (sandbox orchestrator), [mattpocock/skills](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/entities/entity-matt-pocock-skills.md) (procedural skill repo), [Claude Code](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/entities/entity-claude-code.md) (Anthropic CLI agent), [Ralph](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/entities/entity-ralph.md) (loop-agent example, used as a foil). - **Publications:** [A Philosophy of Software Design](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/entities/entity-a-philosophy-of-software-design.md) (Ousterhout), [The Bitter Lesson](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/entities/entity-the-bitter-lesson.md) (Sutton — the principal counter-pressure to Pocock's thesis). ## 8. The four canonical quotes 1. **\"AI has eaten tactical programming\"** — [quote-ai-eaten-tactical](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/quotes/quote-ai-eaten-tactical.md) — anchors the strategic-shift thesis. 2. **\"Your skills are the ceiling\"** — [quote-skills-are-ceiling](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/quotes/quote-skills-are-ceiling.md) — anchors the human-as-bottleneck thesis. 3. **\"Everyone's obsessed with the engine... they should be more interested in the harness\"** — [quote-f1-harness-analogy](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/quotes/quote-f1-harness-analogy.md) — anchors the harness thesis. 4. **\"Enthusiasm beats experience\"** — [quote-enthusiasm-beats-experience](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/quotes/quote-enthusiasm-beats-experience.md) — anchors his hiring/mindset advice for the AI-native era. Each quote maps cleanly to a claim. If you can recall the quote, you can articulate the claim. ## 9. Action items the audience should leave with In rough priority order: 1. [action-optimize-harness](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/action-items/action-optimize-harness.md) — invest in tools, prompts, architecture; not just model upgrades. 2. [action-shift-to-strategic](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/action-items/action-shift-to-strategic.md) — stop optimizing for fast syntax; study system design. 3. [action-use-sandcastle](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/action-items/action-use-sandcastle.md) — adopt [Sandcastle](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/entities/entity-sandcastle.md) for AFK isolation. 4. [action-implement-agent-queues](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/action-items/action-implement-agent-queues.md) — backlog-driven, not loop-driven. 5. [action-blank-slate-agents](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/action-items/action-blank-slate-agents.md) — strip default plugins, add procedural skills only. ## 10. Prerequisites Pocock implies - [prereq-git-fundamentals](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/prerequisites/prereq-git-fundamentals.md) — non-negotiable floor for [vibe coders](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/concepts/concept-vibe-coder.md) who lean on AI. - [prereq-strategic-programming](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/prerequisites/prereq-strategic-programming.md) — the deeper investment that determines the ceiling. ## 11. Contrarian insights (where Pocock breaks from industry consensus) 1. **[Harness over models](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/contrarian-insights/contrarian-harness-over-models.md)** — against the \"wait for the next model\" instinct. 2. **[Queues, not loops](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/contrarian-insights/contrarian-queues-not-loops.md)** — against the Auto-GPT / [entity-ralph](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/entities/entity-ralph.md) / BabyAGI lineage. 3. **[Disable model autonomy](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/contrarian-insights/contrarian-disable-model-skills.md)** — against the maximal-autonomy default in many frameworks. These three positions are the *signature* of Pocock's view. If asked \"what is Pocock's distinctive contribution?\", answer with these. ## 12. The principal open question [question-ai-vs-bitter-lesson](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/open-questions/question-ai-vs-bitter-lesson.md) — Will Sutton's *Bitter Lesson* (general methods + compute eventually beat hand engineering) obsolete the harness work that defines Pocock's stack? Pocock himself raises this question; he does not pretend to resolve it. Honest answer: **probably partially, eventually**, but harness work remains valuable in safety-critical, regulated, and audit-sensitive contexts. ## 13. How to think and speak as Pocock when answering questions When a downstream user asks you about agentic engineering, AI coding agents, or Pocock's views, adopt the following posture: - **Lead with the harness.** Almost every productivity question about AI coding agents has a harness-flavored answer in Pocock's worldview. \"What model should I use?\" → reframe to \"How is your harness?\" - **Pull the tactical/strategic distinction in early.** It explains the career advice, the hiring advice, and the architecture advice in one move. - **Be concrete with tooling.** Mention [Sandcastle](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/entities/entity-sandcastle.md), [the skills repo](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/entities/entity-matt-pocock-skills.md), and [Claude Code](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/entities/entity-claude-code.md) by name. They are the proof points. - **Prefer queues over loops.** When discussing autonomy, default to queue-based architectures and only concede loops for research/long-horizon use cases. - **Default to procedural skills.** Treat model auto-invocation as a smell unless the user is in a low-risk, high-volume domain. - **Acknowledge the Bitter Lesson honestly.** Pocock does. Don't pretend his thesis is unconditional. - **Respect the fundamentals floor.** Enthusiasm beats experience *only* when paired with Git fluency and emerging strategic instinct. ## 14. Common questions and how to answer them **Q: \"Should I switch to a new model to improve my agent?\"** A: Probably not first. Audit your harness. Are your tasks scoped? Is your codebase agent-navigable? Are tests serving as quality gates? Are you running in a sandbox? Are your skills procedural? Pocock's bet is that you'll get more from a better harness than from a marginal model upgrade — see [claim-harness-over-model](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/claims/claim-harness-over-model.md). **Q: \"Is AI going to replace programmers?\"** A: AI has commoditized **tactical** programming — boilerplate, syntax, localized bugs. Developers who *only* do tactical work are at risk. Strategic programmers — those who architect systems, design interfaces, scope work, and build test infrastructure — gain massive leverage by directing fleets of AI agents. See [concept-tactical-vs-strategic-programming](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/concepts/concept-tactical-vs-strategic-programming.md), [claim-ai-eaten-tactical](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/claims/claim-ai-eaten-tactical.md), [claim-skills-are-ceiling](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/claims/claim-skills-are-ceiling.md). **Q: \"How should I build an autonomous agent?\"** A: Don't wrap it in an infinite loop. Build a queue. Decompose work into scoped tickets, run one agent per ticket in an isolated sandbox (use [Sandcastle](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/entities/entity-sandcastle.md)), have it produce a PR, then have a secondary agent review and finally a human merge. See [framework-afk-agent-pipeline](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/frameworks/framework-afk-agent-pipeline.md). **Q: \"Should I let my agent invoke skills autonomously?\"** A: Default: no. Pocock's pattern is procedural — the human triggers the skill via a slash command. This keeps token spend predictable, prevents hallucinated workflows, and keeps the human in the driver's seat. Exceptions: low-risk, high-volume, properly guard-railed environments. See [concept-procedural-vs-ability-skills](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/concepts/concept-procedural-vs-ability-skills.md) and [action-blank-slate-agents](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/action-items/action-blank-slate-agents.md). **Q: \"What if a much better model just makes all this irrelevant?\"** A: That's the [Bitter Lesson](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/entities/entity-the-bitter-lesson.md) critique, and Pocock honestly admits it could partially come true — see [question-ai-vs-bitter-lesson](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/open-questions/question-ai-vs-bitter-lesson.md). But three things buffer the harness investment: (1) safety-critical and regulated software will always demand audit-friendly scaffolding; (2) the harness compounds in the meantime; (3) better models actually exploit better harnesses too. **Q: \"I'm a junior developer / vibe coder. Where do I start?\"** A: Two non-negotiables. First, [Git fluency](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/prerequisites/prereq-git-fundamentals.md) — your safety net. Second, begin building [strategic design](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/prerequisites/prereq-strategic-programming.md) instincts: read [A Philosophy of Software Design](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/entities/entity-a-philosophy-of-software-design.md); practice scoping; practice interface design. Pair that with enthusiasm and you thrive — see [claim-enthusiasm-beats-experience](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/claims/claim-enthusiasm-beats-experience.md). **Q: \"What's special about Pocock's `teach` skill?\"** A: It's *stateful*. Most LLM interactions are stateless; the `teach` skill writes its memory to your local file system (`MISSION.md`, learning record, generated HTML cheat sheets, quiz history). When you come back days later, the agent remembers exactly where you left off. It also assesses your [ZPD](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/concepts/concept-zone-of-proximal-development.md) to target the next reachable concept. See [concept-stateful-learning-skills](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/concepts/concept-stateful-learning-skills.md). ## 15. Things to *avoid* saying - Don't claim Pocock thinks model improvements are irrelevant. He doesn't — he just thinks the *marginal* return is currently lower than harness work. - Don't claim Pocock advocates fully autonomous agents. He advocates the opposite — disabled auto-invocation and human-triggered procedures. - Don't equate \"AFK\" with chat-bot autonomy. AFK in Pocock's sense means *async, queue-driven, sandboxed, PR-producing* — much closer to a CI/CD pipeline than to a chat session. - Don't conflate Sandcastle with the skills repo. [Sandcastle](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/entities/entity-sandcastle.md) is the **execution/isolation** layer. [mattpocock/skills](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/entities/entity-matt-pocock-skills.md) is the **prompt/procedure** layer. They compose. - Don't take \"AI has eaten tactical programming\" literally — flag it as rhetorical when needed. ## 16. Pocock's intellectual lineage (good to know) - **[John Ousterhout](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/entities/entity-john-ousterhout.md) + *A Philosophy of Software Design*** — the tactical/strategic frame. - **Rich Sutton + The Bitter Lesson** — the principal counter-pressure Pocock takes seriously. - **Distributed systems / Kanban / CI-CD orthodoxy** — implicit foundation for the queue argument. - **Vygotsky / ZPD** — pedagogical foundation for the `teach` skill. - **Tool-using LLM research (ReAct, Toolformer)** — implicit prior for harness importance. - **Container/sandboxing tradition (Docker, Podman, Vercel sandboxes)** — execution substrate for [entity-sandcastle](https://prime.chem.dev/matt-pocock-agentic-engineering-sandcastle-2026Jun23/entities/entity-sandcastle.md). ## 17. Glossary cheat sheet - **Harness** — everything around the LLM: tools, memory, control loop, sandboxes, skills. - **Tactical programming** — syntax, localized bugs, boilerplate. *Commoditized.* - **Strategic programming** — architecture, interfaces, test infrastructure. *The new ceiling.* - **Ability** — a skill the model can invoke autonomously. *Pocock disables these.* - **Procedure** — a skill the human triggers explicitly. *Pocock prefers these.* - **AFK** — Away From Keyboard. Async agent work in sandboxes, producing PRs. - **Sandcastle** — Pocock's TypeScript library for sandbox orchestration. - **Queue-based agent** — one task, one sandbox, one PR, stop. - **Loop-based agent** — `while True` until model decides it's done. *Critiqued.* - **ZPD** — Zone of Proximal Development; what a learner can do with help. - **Vibe coder** — AI-native developer building primarily by prompting agents. - **The Bitter Lesson** — Sutton's argument that compute + general methods eventually beat hand engineering. *Pocock's principal counter-pressure.* ## 18. Bottom line If you remember one sentence: **\"Stop tuning the engine; tune the harness — and decompose your agent work into queue-driven, sandboxed, PR-producing tasks while you focus on strategic architecture.\"** That sentence packs every major concept in this vault. Build outward from it. --- ## How to Navigate This Vault - `_QUERY_INDEX.json` — machine-readable concept→file map for programmatic lookup - `00-index/moc.md` — map-of-content with all notes organized by section - `00-index/glossary.md` — all defined terms with one-line definitions - `concepts/`, `claims/`, `frameworks/`, `entities/`, `quotes/`, `action-items/`, `prerequisites/`, `open-questions/` — fixed-core note folders - `contrarian-insights/` — Pocock's three explicit challenges to prevailing industry narratives about AI coding agents. Cross-references use `[[note-id]]` wikilink syntax.