Verify CLAUDE.md Imports Actually Loaded with /context
There's a common mistake when setting up CLAUDE.md imports: writing the filename as plain text does nothing. Claude sees it as text, not as a file to load.
# Wrong — Claude reads "AGENTS.md" as plain text
AGENTS.md
# Correct — Claude loads and injects the file contents
@AGENTS.md
The @ prefix is what triggers the import. Without it, the referenced file is completely ignored.
The easiest way to confirm your imports are working is the /context command:
/context
This shows every file currently loaded into Claude's context window. If a file you imported isn't listed, it hasn't been loaded — check for the missing @ prefix, or verify the path is correct relative to the CLAUDE.md file location.
Common import patterns that require the @ syntax:
# Load a shared team instructions file
@docs/team-conventions.md
# Pull in the project README for background context
@README.md
# Import a separate agents config
@AGENTS.md
A quick diagnostic workflow: add @yourfile.md to CLAUDE.md, start a new session, run /context, and confirm the file appears. If it doesn't, the path is wrong or you forgot the @.
No @, no import — and /context is how you prove it worked.
Log in to leave a comment.
When Claude writes error messages, button labels, validation text, or onboarding flows, it defaults to generic developer-speak. Add a "Users" section to your CLAUDE.md describing who your actual users are — their technical level, industry jargon, and what they care about — so Claude writes copy that makes sense to THEM, not to developers.
Use the --agent flag with custom markdown files in .claude/agents/ to launch purpose-built Claude sessions with restricted tools and scoped system prompts.
Every project has traps — the billing module that silently fails if you forget to queue the job, the legacy table with column names that don't match the model, the config value that must be set before tests run. Document these gotchas in your CLAUDE.md so Claude avoids the same mistakes your team spent days debugging.