daledesilva/obsidian_ink / .devin/skills/debug-ingest
debug-ingest
VerifiedDebug ingest (Wi‑Fi / Cursor Debug)
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 3.1 KB |
SKILL.md
View on GitHub---
name: debug-ingest
description: Debug ingest (Wi‑Fi / Cursor Debug)
triggers:
- user
allowed-tools:
- read
- exec
- write
- edit
- mcp_call_tool
---
# Debug ingest (Wi‑Fi / Cursor Debug)
Set up live NDJSON ingest from the Obsidian plugin (desktop or mobile) into Cursor’s Debug log file so the agent can watch runtime evidence.
Any text after `/debug-ingest` is optional context (e.g. iPad, session id).
## Preconditions
- User should be in (or about to start) a **Cursor Debug** agent session — shell scripts **cannot** start the ingest listener.
- Note from Debug session context: **session ID** (slug), **ingest path** (`/ingest/<uuid>`), **ingest port**, **log file** (`.cursor/debug-<sessionId>.log`). Path UUID ≠ session slug.
## Steps
1. **Confirm Debug ingest is live**
- If this is not a Debug-mode session, tell the user to start one (or switch) before continuing.
- Record session id, ingest path, ingest port, and log file path from the system/debug context.
2. **Start the LAN relay** (required for a device that is not this Mac):
```bash
bash scripts/ingest-lan-relay.sh
```
Run in the background. Requires `socat`. It forwards `0.0.0.0:7662` to localhost. Ink does not read the offer on port 7663.
3. **Bake the session into a local build**, then copy `dist/`:
```bash
INK_DEBUG_CURSOR_SESSION_ID=<sessionId> \
INK_DEBUG_INGEST_PATH=<ingestPath> \
npm run build
```
For iPad or another computer, also set `INK_DEBUG_LAN_IPV4` to this Mac's Wi-Fi address. Never bake `127.0.0.1` for a remote device. Copy `dist/main.js`, `dist/styles.css`, and `dist/manifest.json` into the vault's `.obsidian/plugins/ink/`. `npm run internal-release` builds committed code only.
4. **Optional offer** (the plugin does not request it):
```bash
bash scripts/write-debug-ingest-offer.sh \
--session <sessionId> --path <ingestPath> --port <ingestPort>
python3 scripts/debug-ingest-offer-sidecar.py
```
Host check: `bash scripts/print-debug-ingest-settings.sh`.
5. **Clear only the session Cursor Debug log** when preparing a **new** reproduction you are about to ask for (not when analyzing a run the user already finished):
- Truncate/delete **only** `.cursor/debug-<sessionId>.log` for this session.
- Do **not** delete other sessions' logs. See `.cursor/rules/preserve-sync-debug-logs.mdc`.
6. **Reproduce**, then **read** `.cursor/debug-<sessionId>.log` and cite line evidence before fixing.
7. After a verified fix, remove temporary ad-hoc ingest call sites. Keep `src/logic/utils/cursor-debug-ingest.ts`.
## Do not
- Use `fetch('http://127.0.0.1:…')` inside the plugin — use `requestUrl` via `postCursorDebugIngest`.
- Assume USB alone bridges iOS to localhost (use Wi‑Fi + relay).
- Treat silent ingest failure as proof a code path did not run — read `.cursor/debug-<sessionId>.log` after a reproduction.
- Treat the offer sidecar as the ingest path. Ink posts to the baked URL through `scripts/ingest-lan-relay.sh`.
See `docs/debugging-lan-ingest.md`, `docs/debugging-on-ipad.md`, and `.cursor/rules/cursor-debug-ingest.mdc`.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| finishing-a-development-branchobra/superpowersUse when implementation is complete, all tests pass, and you need to decide how to integrate the work | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| systematic-debuggingobra/superpowersUse when encountering any bug, test failure, or unexpected behavior, before proposing fixes | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| test-driven-developmentobra/superpowersUse when implementing any feature or bugfix, before writing implementation code | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| skill-creatoranthropics/skillsCreate new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |