A Codex Skill is a reusable folder of instructions — a SKILL.md file plus optional scripts and reference docs — that teaches Codex how to perform a specific, repeatable job, like reviewing Oracle PL/SQL code the exact way your team wants it reviewed, every single time. Without a Skill, every PL/SQL review starts from zero: you re-explain "check for unhandled exceptions, check for row-by-row processing, check for SQL injection in dynamic SQL" in every single prompt, and Codex's review quality depends entirely on how well you happened to phrase that prompt that day. A Skill fixes this permanently — write the review checklist once, and Codex applies it consistently, whether it's you, a teammate, or a CI job invoking it. 📦 A Skill is just a folder — SKILL.md plus optional scripts/ , references/ , and assets/ 🎯 Skills activate two ways: explicitly (you type $skill-name ) or implicitly (Codex matches your request to the skill's description) 🏗️ Codex loa...

The Model Context Protocol (MCP) is an open standard, introduced by Anthropic in November 2024, that gives AI assistants like Codex a single, standardized way to discover and call external tools — including databases — instead of relying on one custom integration per tool. 🔌 Why this matters in a corporate setting: as of mid-2026, roughly 78% of enterprise AI teams already run MCP-backed agents in production , and about 28% of Fortune 500 companies operate their own MCP servers — Bloomberg alone used MCP to close what its engineering team calls the "productionization gap" across more than 9,500 engineers , turning one-off GenAI demos into governed, deployable systems. If you're learning to connect an AI coding assistant to a production database today, you are learning the exact connective layer enterprises are standardizing on right now. 🏢 📑 In This Post What MCP Actually Is (And Why It Exists) Why AI Assistants Need MCP To Reach Databases Meet the P...

Yes — Codex loads every installed skill's name and description into context at startup, and that costs real tokens; allow_implicit_invocation: false is the setting that removes a specific skill from that startup list entirely, so it stops costing tokens and stops auto-triggering, while still staying reachable if you type its name directly. 🎯 This is worth getting precise about, because the intuitive guess is usually wrong on one detail: people assume this flag just "stops the skill from firing automatically," and stop there. What actually happens is stronger — the skill's description is pulled out of the model-visible list altogether, not merely muted. That distinction matters the moment you have a sensitive or expensive skill (one that touches production data, spends a lot of tokens, or should only run when a human explicitly asks for it) sitting alongside dozens of others. 🔐 The same skill, two different footprints, depending on one policy field. 1....

SKILL.md is the file format behind Agent Skills — a folder containing a single Markdown file with YAML frontmatter (a name and a description) plus a body of plain-English instructions, which an AI agent like Claude loads on demand to gain a specific capability it didn't have a moment ago. 🗂️ This matters because the old way of teaching an agent your team's conventions was to cram everything into one always-loaded prompt — and that approach collapses the moment you have more than a handful of workflows to teach it. A always-loaded instructions file that covers forty different situations burns context on the thirty-nine that don't apply to the current task, crowding out the model's actual reasoning room and degrading answer quality as the file grows. SKILL.md solves this with progressive disclosure: an agent can have dozens of skills installed and pay almost nothing for the ones it isn't using right now. Anthropic published the format as an open standard, and it...