claude-api Skill Adds a Non-Interactive Prompt-Audit Subcommand
Anthropic's claude-api Claude Agent Skill gained a prompt-audit subcommand that scans a prompt's instructions against dated-pattern tables and outputs a findings report with a proposed diff.
What it is
A new non-interactive subcommand added to Anthropic's claude-api skill, the reference skill Claude Code uses for Anthropic API guidance.
What it does
It inventories a prompt's instruction surface, dates each instruction, checks them against known dated-pattern tables for stale or outdated guidance, and produces a findings report plus a proposed diff, all without requiring interactive back-and-forth. The skill's model-migration guidance now points to this audit from its prompt-behavior section.
Why it matters
Prompts accumulate stale instructions the same way codebases accumulate dead code, and this gives teams a repeatable way to check for drift instead of manually re-reading every system prompt after each model migration.
How to use it
Invoke the claude-api skill's prompt-audit subcommand against a target prompt to get a dated inventory and a suggested diff for stale instructions.
A 1,024-Character Limit Just Forced a Skill Rewrite
Anthropic renamed its claude-academy-guide Claude Agent Skill to academy-guide and shortened its description to comply with skill-upload validation rules for custom skills.
claude-api Skill Gets an `upgrade` Subcommand for the 0.x to 1.x SDK Jump
Anthropic's claude-api Claude Agent Skill added an upgrade subcommand that walks a Python project through migrating from the anthropic 0.x SDK to 1.x, including the new httpx2 HTTP layer.
Claude Academy Guide Skill Reads a Live Course Catalog So It Never Goes Stale
The academy-guide Claude Agent Skill recommends matching Claude Academy courses and tutorials by reading a live catalog JSON at runtime instead of an embedded snapshot.