[CompleteDotTech]

Resolves #132

Change and affected skills

Publishes the skill's own normalized icon (assets/icon.png, 54099 bytes), its regenerated banner and card (docs/images/banner.png, docs/images/card.svg) and the docs/images/.showcase.json sidecar, all produced through the real showcase.py from freshly fetched public metadata. The key chip "Claude + Codex" and the attribute chips (Python, Offline) are persisted in .github/showcase.yml (chips.key, chips.attrs), so normal regeneration keeps them and the card no longer repeats the runtime chip next to the key chip. The exact icon provenance is recorded in assets/PROMPTS.md: the recorded adapted prompt, the original generation on 2026-10-31 versus the normalization and regeneration on 2026-11-01, and the tool described as OpenAI built-in image generation invoked by Codex (model ID not exposed, seed unsupported, no CLI transport claimed). No change to scripts/, references/, action.yml or templates.

Prior work

Searched open and closed issues and PRs for key chip, showcase.yml key, regenerate assets, icon provenance and banner card social own: only #98 is adjacent and not duplicated; merged #130, #126, #129 are reused. No other PR covers this branch or issue.

Evidence (head a6ba563bce2d7304736f80dd4bc84a8555ee45ff, base main 04fbc0aff8320fa58c16422009479fdb75d84ba2)

• Exact-head VM: all four required commands (package tests, package validator, full pytest, quality validator) passed in the designated VM: 1307 passed, 11 skipped, 37 subtests passed. Native Windows full suite (supplemental): 1300 passed, 18 skipped. Whole-tree ruff passes.
• Tests: tests/test_own_config_chips.py is a synthetic, no-network regression showing that the configured chips override the runtime chip with no duplicate.
• Final replay, no --icon and no chip flags, from the committed configuration with the same pinned public metadata: a native Windows replay reproduces all six approved files byte for byte (banner, card SVG and PNG, social image, both contact sheets). The VM replay produces the identical SVG and identical decoded pixels (RGBA and decompressed image data) for every PNG, but not identical PNG compressed bytes (Windows zlib-ng 1.3.1 against Linux zlib 1.3); this PR does not claim that VM PNG bytes equal the committed files.
• Visual review (Codex role, root; visual-review-132-a6ba563.json): PASS. Both contact sheets were viewed on the dark and light GitHub backgrounds at 1012 and 390 wide: the key chip "Claude + Codex" appears once, the blue Python and Offline chips fit without duplication, and there is no overlap or clipping. A representative public org banner, tile and intro gallery also passed the CLI and SVG lints and manual viewing.
• Original optional-renderer and Windows logging failures during the replays are preserved in the evidence and the successful retries are recorded.
• The committed assets are byte-identical copies of the visually accepted render; the raw Codex image and the private metadata file are not committed. Codex remains the image owner and nothing was hand-edited.
• Hosted CI on this exact head: all 14 required check runs (validate-package, pytest on ubuntu/windows/macos, validate; push and pull_request) completed with success; the PR is mergeable and clean against main with zero review threads.
• Reviews on this exact head and base, all approve with no findings and the source criteria met: two independent Opus reviews (both claude-opus-5-5, native identities verified from the agent transcripts) and one Codex gpt-6.1-sol review at max effort (native Codex session verified; the lifecycle gates are coordinator-owned and were checked by the coordinator). Earlier malformed review receipts were not selected.

Boundaries

No activation keys, consent, private assets, approval gates or specialist entry points changed. No registry release, visibility change or live external action is implied. The repository-presentation audit (settings, labels, health, social preview) and the historical example-sample alignment are separate backlog items and not part of this PR.

Family consistency

No shared change; no sibling PRs affected.

🤖 Generated with Claude Code

[chatgpt-codex-connector]

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	⚠️ Failed 2026-11-01T09:45:27.056182Z We couldn’t complete this review. Comment @codex review to try again.	a6ba563	PR opened
🔒 Security Review	⚠️ Failed 2026-11-01T09:45:33.935370Z We couldn’t complete this review. Comment @codex security review to try again.	a6ba563	PR opened

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

[chatgpt-codex-connector]

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.
To continue using code reviews, add credits to your account and enable them for code reviews in your settings.

[chatgpt-codex-connector]

The account paying for this security review has reached its Codex usage limits. The payer can check the Codex usage dashboard. For personal accounts, using credits requires enabling “Use credits for security reviews” in Code review settings. If you do not manage the paying account, contact this repository's admins.