Repository navigation
feat: af_load_data skill — own data loading and selection cuts - #62
Merged
Merged
Conversation
New skill that loads the user's data (CSV / FITS / HDF5 / npy / JSON) into the arrays their Analysis takes, confirms column meanings and units, states every selection cut back with its justification, and records both in wiki/project/ before the data-inspection gate fires. Registered in the skills index, .claude and .codex discovery; cited from af_adapt_to_domain, af_wrap_likelihood and start-here P0; profile template gains "Data shapes". Closes #61 Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_016nhQ8N1DRK5UkzPPm3tuZQ
Adds the missing .claude/commands/af_load_data.md symlink and refreshes the stale af-configure-search Codex adapter description (pre-existing drift on main), via PyAutoBrain/bin/project_discovery.py write. install.sh --check-project-discovery autofit_assistant now passes. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_016nhQ8QHqrNT4BEqeURYq8y
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Closes #61.
Summary
skills/af_load_data.md— reads CSV / FITS / HDF5 / npy / JSON into the arrays the Analysis takes; asks the user to confirm column meanings, units and error convention; states every selection cut back with a one-line justification and prints rows kept after each cut; records the column map in profile "Data shapes" and the cuts in a datedwiki/project/entry; hands off to the data-inspection gate immediately. Worked recipe is the bundled Pantheon+ CSV (is_calibrator == 0,z_hd > 0.023), written in workspace script style.skills/README.md(Domain adaptation),.claude/skills/af_load_data.mdsymlink,.codex/skills/autofit-assistant-af-load-data/SKILL.md.af_adapt_to_domain(interview step 2 + Orient table),af_wrap_likelihood(Ask: the data),modes/start_here.mdP0 now routes loading through the skill instead of the one-sentence stopgap.wiki/project/_profile_template.mdgains a "Data shapes" section.Notes for review
.codexadapter was written by hand to match the formatPyAutoBrain/bin/project_discovery.pyproduces: PyAutoBrain could not be attached to this session, so the generator wasn't run. It is byte-identical in structure to the existing adapters; please runinstall.sh --check-project-discovery autofit_assistantbefore merging (or regenerate).is_calibrator == 0→ 1371 afterz_hd > 0.023, all finite.pytest autoassistant/tests: 13 failures, the same 13 as unchangedmainin this container (autofit isn't installed, plus one existing benchmark README check). None come from this change.Generated by Claude Code