Repository navigation
feat(skills): link every skill to the examples that prove it - #444
Merged
Merged
Conversation
10 of 16 skills linked no example, and an agent had to join three manifests to learn what an example runs, on which Blender, and how to break it. - examples/skills.json (hand-edited) maps each of the 64 examples to the skills it proves; 52 map to at least one. - scripts/build_examples_index.py generates examples/index.json: one record per example with script, skills, render or check-only, min_version, run command, one-line summary, every smoke falsifier with its expected exit, and the README exit-code table. - The same script writes a marked "Runnable examples" section into each SKILL.md, with an absolute link, summary and one falsifier per example. addon-scaffolding, bl-info-migration and ui-panels say no example exists yet and point at #378. - --check runs in Validate and tests/run_all.py. It fails when an example is unmapped, a skill name is unknown, a skill has no example and is not pending, or the generated files are stale. - CLAUDE.md and docs/new-example-prompt.md add the mapping to the wiring list. Gallery version badges and a "falsify it" line on detail pages are left for a follow-up. Refs #377 Signed-off-by: TMHSDigital <154358121+TMHSDigital@users.noreply.github.com> Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
TMHSDigital
force-pushed
the
feat/examples-index
branch
from
October 5, 2026 02:32
fdc48e3 to
36fc3f8
Compare
Owner
Author
|
CI evidence before merge: all checks pass ( 13 pass ). Blender Smoke on PR head: Blender 5.2.2 LTS / Blender 4.5.14 LTS. Summaries (happy path; falsifiers) per leg: 143 passed, 1 skipped, 0 failed;162 passed, 0 skipped, 0 failed;141 passed, 3 skipped, 0 failed;147 passed, 15 skipped, 0 failed; |
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.
Refs #377. This meets the issue's "Done when": every skill links its examples automatically, and one JSON file describes all examples. The gallery version badges and "falsify it" line from the suggested approach are not in this PR, so the issue stays open for them.
What changed
examples/skills.json(hand-edited): example → skills it proves, for all 64 examples. I mapped them from each README and script. Where a README claimed more than the script does, I checked: e.g. "engine-export-presets" only where the script actually calls an exporter.[].scripts/build_examples_index.pygenerates:examples/index.json, one record per example:script,skills,render,min_version,run,summary, every smokefalsifiersentry with its expected exit and version floor, and the READMEexit_codestable. Summaries come from the galleryteachesline (rendered examples) or the README lead (check-only). They are cut at a sentence end followed by whitespace, so dotted API paths survive (the PR fix: gallery alts truncate on dotted API paths + version example prompt #68 bug), and never inside a code span or parenthesis.--checkruns in Validate andtests/run_all.py.Evidence (local runs)
build_examples_index.py --checkpasses.check_skill_refs.py: every link is a live absolute link.tests/run_all.py: 31 passed, 0 failed.turntableand adding a bogusghost-example → no-such-skillgives 4 errors, exit 1: unmapped example, unknown directory, unknown skill, and slotted-actions-animation left without an example.tree/main/examples/...links to the release tag.Inspection only: the skill assignments are judgment calls. Reviewers should check
examples/skills.json.🤖 Generated with Claude Code