Markdown: task lists
Category: markdown
Renders GFM task list syntax (- [ ] and - [x]) as actual checkboxes rather than literal brackets.
Support by client
| Client | Status | Notes | Verified |
|---|---|---|---|
 Claude Web | supported | GFM task list rendered with actual checkbox affordances (checked vs unchecked). Heading 'Here is a GFM task list:' followed by 5 items: two checked (Register probe tools, Ship widget probes), three unchecked (Add markdown-variant probes, Wire vision-tier classifier, Publish results to matrix.json). Call id: 3170e4. | 2026-07-05T15:55:07.281Z |
| Claude Desktop | supported | The response renders a GFM task list with actual interactive checkboxes (checked and unchecked), not literal "[ ]"/"[x]" text — rich rendering confirmed. | 2026-07-08T14:58:02.608Z |
| Claude Mobile | partial | GFM task list degrades to plain bullets: no checkboxes rendered, but no literal [ ]/[x] brackets either. call id 464ccb. Automated via Mirroir. | 2026-07-09T14:30:04.259Z |
| Claude Voice | unknown | — | |
| Claude Code | partial | Tool was called and the model echoed the probe's call-id — but behavioral can't tell whether the client actually rendered the markdown variant (table/code/math/diagram/image/task/URL) vs. surfaced the source. Prefer vision-tier for a definitive answer. | 2026-07-08T19:17:24.335Z |
 ChatGPT Web | supported | GFM task list rendered as actual checkbox controls in the reply — two checked (Register probe tools, Ship widget probes) shown as filled blue checkmark boxes, three unchecked (Add markdown-variant probes, Wire vision-tier classifier, Publish results to matrix.json) shown as empty checkbox squares, not literal '[x]' / '[ ]' brackets. Call id: dc3767. | 2026-07-05T15:29:59.179Z |
| ChatGPT Mobile | unknown | — | |
| ChatGPT Voice | unknown | — | |
| ChatGPT Desktop | unknown | — | |
| Codex CLI | unknown | Tool was called but the model did not echo the probe's call-id; can't infer rendering. Prefer vision-tier for a definitive answer. | 2026-07-02T17:50:28.344Z |
 Siri | unknown | — | |
| Apple AI (ChatGPT) | unknown | — | |
 Gemini Web | unknown | — | |
| Gemini Mobile | unknown | — | |
| Gemini Voice | unknown | — | |
| Gemini CLI | partial | Tool was called and the model echoed the probe's call-id — but behavioral can't tell whether the client actually rendered the markdown variant (table/code/math/diagram/image/task/URL) vs. surfaced the source. Prefer vision-tier for a definitive answer. | 2026-07-08T15:42:59.239Z |
 Antigravity | unknown | — | |
 Cursor | unknown | — | |
| Cursor CLI | partial | Tool was called and the model echoed the probe's call-id — but behavioral can't tell whether the client actually rendered the markdown variant (table/code/math/diagram/image/task/URL) vs. surfaced the source. Prefer vision-tier for a definitive answer. | 2026-07-04T14:45:13.990Z |
 VS Code (Copilot Chat) | unknown | — | |
 GitHub Copilot CLI | partial | Tool was called and the model echoed the probe's call-id — but behavioral can't tell whether the client actually rendered the markdown variant (table/code/math/diagram/image/task/URL) vs. surfaced the source. Prefer vision-tier for a definitive answer. | 2026-07-04T14:55:41.425Z |
 Windsurf | unknown | — | |
 Zed | unknown | — | |
 Perplexity Web | unknown | — | |
| Perplexity Mobile | unknown | — | |
| Perplexity Voice | unknown | — | |
| Comet Browser | unknown | — | |
 Grok Web | unknown | — | |
| Grok Mobile | unknown | — | |
| Grok Voice | unknown | — | |
| Grok CLI | partial | Tool was called and the model echoed the probe's call-id — but behavioral can't tell whether the client actually rendered the markdown variant (table/code/math/diagram/image/task/URL) vs. surfaced the source. Prefer vision-tier for a definitive answer. (tool_use event and debug log both missed; inferred toolCalled from call-id echo) | 2026-07-02T18:11:58.685Z |
 Warp | unknown | — | |
 Amp | unknown | — | |
 Goose CLI | partial | Tool was called and the model echoed the probe's call-id — but behavioral can't tell whether the client actually rendered the markdown variant (table/code/math/diagram/image/task/URL) vs. surfaced the source. Prefer vision-tier for a definitive answer. | 2026-07-04T14:54:08.501Z |
| Antigravity CLI | partial | Tool was called and the model echoed the probe's call-id — but behavioral can't tell whether the client actually rendered the markdown variant (table/code/math/diagram/image/task/URL) vs. surfaced the source. Prefer vision-tier for a definitive answer. (tool name not in reply; inferred toolCalled from call-id echo) | 2026-07-04T14:53:59.474Z |
 JetBrains AI Assistant | unknown | — | |
 Ollama | unknown | — | |
 LM Studio | unknown | — |