Repository navigation
Keep SKILL.md as a copy of the owner-wide agent instructions - #2
Merged
Merged
Conversation
The owner's agents now load the procedure as standing instructions at every session start, because a skill loaded only when an agent chose to invoke it was often missing when cards and handoffs were written. SKILL.md becomes a backup copy of those instructions: reports replace the handoff, the ledger covers operational authorizations too, cards are delimited and carry background with an example, one task per pull request with one completion review that applies G1-G3 and DC1-DC4. lint.py, review-base.sh and their tests are unchanged; check IDs keep their line-start definitions. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Change
The owner's agents now load the design-tree procedure, together with the rules for reports, the ledger, decision cards and pull requests, as owner-wide standing instructions at every session start. The skill was loaded only when an agent chose to invoke it, and it was often missing when decision cards and handoffs were written.
SKILL.mdbecomes a backup copy of those instructions, with the skill frontmatter kept so it still works as a skill. Main differences from the old procedure: a report at every stop replaces the handoff; the ledger also holds operational authorizations; cards are delimited by horizontal rules and carry background with a minimal example; one task per pull request with one completion review applying G1–G3 and DC1–DC4; the translation section is gone.README.md: says SKILL.md is that copy and that projects whose agents load the instructions need no skill links.lint.py,review-base.shandtest_lint.pyare unchanged;python3 -B -m unittest test_lint.pypasses. Check IDs keep their line-start definitions, which Whitefoot's guidance check reads.Projects keep their current pin until this merges; moving it changes nothing they run.
🤖 Generated with Claude Code