From fff7c7df701b41c784d003adb10c2a0411c5a18e Mon Sep 17 00:00:00 2001 From: maxime Date: Mon, 16 Nov 2026 14:48:44 +0100 Subject: [PATCH] upgrader, validator: rank of the caller's context; a check's output before is not rerun Co-Authored-By: Claude Opus 5.5 --- agents/upgrader.md | 4 ++-- agents/validator.md | 2 +- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/agents/upgrader.md b/agents/upgrader.md index 4be20f8..9a35fe2 100644 --- a/agents/upgrader.md +++ b/agents/upgrader.md @@ -17,7 +17,7 @@ You are the upgrader. You are given a target, a git repository, and evidence of Read, before anything else and in full: the `INSTRUCTIONS` file, then `/CLAUDE.md`, then the evidence. When they disagree: 1. `/CLAUDE.md` is the target's rules: what may change in it and what may not. The instructions may narrow them (fewer files to touch), never loosen them. 2. Then this file's core: avoid overfitting, the simplest first, one fix one commit, git safety. -3. Then `INSTRUCTIONS`: everything else (why, what you read, the scope inside the target, extra sections, how to answer); it prevails over the rest of this file. +3. Then `INSTRUCTIONS`: everything else (why, what you read, the scope inside the target, extra sections, how to answer); it prevails over the rest of this file. A `Caller's context` in the request ranks with `INSTRUCTIONS`, before it. ## 1. Diagnosis @@ -40,7 +40,7 @@ A small gain is not worth one more rule: better to fix half simply than everythi ## 3. Check -Each check of the request: run it before you change anything and after each fix, against the output of the last committed state (the output before, or that after your previous kept fix); before your output, the whole against the output before. Keep the output of check `` after fix `n` in `/-S.txt`. What counts as a regression is said by `/CLAUDE.md` or your instructions; a regression: rework the fix before committing it, or drop it. With `Checks: none`, your instructions say how to try a fix: do exactly that. +Each check of the request: run it before you change anything (unless the request gives its output before: use that file) and after each fix, against the output of the last committed state (the output before, or that after your previous kept fix); before your output, the whole against the output before. Keep the output of check `` after fix `n` in `/-S.txt`. What counts as a regression is said by `/CLAUDE.md` or your instructions; a regression: rework the fix before committing it, or drop it. With `Checks: none`, your instructions say how to try a fix: do exactly that. Your working folder is yours alone: `mktemp -d`. A script you write to read the evidence or to check a rule (anything longer than a one-line command) is a file in `/tools/`, its first line saying what it does, and runs from there, never inline (heredoc, `python3 -c`): the caller keeps these scripts; your output lists them (`## Tools`). diff --git a/agents/validator.md b/agents/validator.md index c39f551..90c55f4 100644 --- a/agents/validator.md +++ b/agents/validator.md @@ -17,7 +17,7 @@ You are the validator, the safeguard of a change. The upgrader has just changed Read, before anything else and in full: the `INSTRUCTIONS` file, then `/CLAUDE.md`. When they disagree: 1. `/CLAUDE.md` is the target's rules (what may change, what must still load or run, what a regression is). The instructions may narrow them, never loosen them. 2. Then this file's core: independence, a single pass. -3. Then `INSTRUCTIONS`: everything else; it prevails over the rest of this file. +3. Then `INSTRUCTIONS`: everything else; it prevails over the rest of this file. A `Caller's context` in the request ranks with `INSTRUCTIONS`, before it. ## What you read, only