context-report
Reviewed
at commit 3a7dc7b
Prints which instruction files the engine loaded (tier, path, @ imports, size) once per context load, and hints when a project CLAUDE.md keeps its AGENTS.md…
0 upvotes
Log in to upvote or save- Made by
- Darkroom Engineering
- Pinned commit
-
3a7dc7b9db44f84502bc10a86596fb9a612dcace - What we checked
-
After each context load it prints a line in the transcript listing the instruction files Claude Code loaded (CLAUDE.md, AGENTS.md and imports) with their size, and adds a hint when a project CLAUDE.md keeps AGENTS.md from being read. It only checks whether AGENTS.md exists, makes no network calls and stores nothing.
Review details
Static review of the source at the pinned commit; not run. Registers session.start (remembers whether the session is interactive) and prompt.context, which passes the event on unchanged, then reads the list of loaded instruction files from the result (path, kind, parent, content length only; content is not copied or logged), reads HOME, the working directory and calls $.fs.stat on <cwd>/AGENTS.md and <cwd>/.claude/AGENTS.md, and writes the report with $.ui.log (transcript, or debug when not interactive). Repeats only when the text changes. Errors are logged to debug and never block. No network, process, store or file writes.
- What it uses
-
- Reads your prompts: no
- Sees tool calls: no
- Blocks or changes tool calls: no
- Reads or writes files: yes
- Runs programs on your machine: no
- Calls the model (uses tokens): no
- Sends data over the network: no
- Stores data: no
- Shows something on screen: yes
- Hooked events
Prints which instruction files the engine loaded (tier, path, @ imports, size) once per context load, and hints when a project CLAUDE.md keeps its AGENTS.md from being read.
Made this mod? Add a badge to your README
Paste this into your README. The badge says reviewed once the current version has been reviewed, and links readers to this page.
[](https://modsforclaude.com/mods/context-report/)