Ar9av/obsidian-wiki / skills/wiki-capture
wiki-capture
VerifiedTurn the current conversation or finding into a structured permanent wiki note. Use for save/capture/preserve requests; QUICK MODE writes a fast _raw/ note. Supports named-vault routing such as @research save this. This captures current discussion, not external source ingestion.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 17.4 KB |
| references/RAW-FORMAT.md | markdown | 2.8 KB |
references/RAW-FORMAT.md
View on GitHub# Raw File Format Reference
Full specification for `_raw/` files written by `wiki-capture` (quick mode).
These files are designed to be promoted by `/wiki-ingest`.
## Frontmatter
```yaml
---
title: "<Descriptive cluster title>"
category: skills
tags:
- topic/<primary-tech>
- <1-3 additional domain tags from vault taxonomy>
summary: "<1–2 sentences, ≤200 chars — what is this finding about?>"
tier: supporting
related: []
extends: null
contradicts: null
superseded_by: null
capture_source: claude-session
project: "<project name or null>"
base_confidence: 0.75
lifecycle: draft
lifecycle_changed: <YYYY-MM-DD>
provenance:
extracted: 0.85
inferred: 0.15
sources:
- "<project> session (<YYYY-MM-DD>)"
---
```
## Body: Finding Block
For bugs and fixes:
```markdown
## <Finding Title>
**Problem:** <what was non-obvious or broken — be specific about the symptom>
**Root cause:** <why it happened — the underlying mechanism, not just the error message>
**Fix:**
```<lang>
// ❌ before
// ✅ after
```
**Confirmed by:** <build pass / test pass / live app / error disappeared>
```
For gotchas and API quirks (no traditional bug/fix arc):
```markdown
## <Gotcha Title>
**Behavior:** <what surprised the user>
**Explanation:** <why it works this way>
**Workaround / Pattern:** <what to do instead>
**Confirmed by:** <how it was validated>
```
Omit sections that have nothing to say. Add a `**Notes:**` block at the end for caveats,
related edge cases, or follow-up questions.
---
## Provenance + Confidence Calibration
Apply provenance markers inline per the `llm-wiki` convention:
| Marker | When to use |
|---|---|
| *(none)* | Explicitly stated in the conversation |
| `^[inferred]` | Synthesized or generalized beyond what was directly said |
| `^[ambiguous]` | Uncertain, potentially incomplete, or contradicted elsewhere |
Use the table below to set `base_confidence` and the `provenance` split:
| Evidence strength | `extracted` | `inferred` | `base_confidence` |
|---|---|---|---|
| Build error + test pass | 0.90 | 0.10 | 0.80–0.90 |
| Fix applied, appeared to work | 0.75 | 0.25 | 0.70–0.75 |
| Discussed, not fully confirmed | 0.60 | 0.40 | 0.60 |
| Reasoned from a single case | 0.50 | 0.50 | 0.55 |
`extracted + inferred` should sum to 1.0 (or include a small `ambiguous` fraction if applicable).
---
## Multiple Findings in One File
When several related findings belong to the same topic cluster, place them sequentially
in the body — each as its own finding block with a `##` heading. Use a short intro paragraph
before the first block to explain what ties them together.
```markdown
# Swift 6 Concurrency Gotchas
Findings from migrating an iOS app to strict concurrency checking. All confirmed during
the build phase.
## Actor reentrancy in async forEach
**Problem:** ...
## MainActor isolation not inferred on @Observable
**Problem:** ...
```
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| add-product-e2e-evalpaperclipai/paperclipAdd or extend a Paperclip full-stack runner E2E workflow, fixture, matcher, or report evidence path for local or Daytona execution. | Stars: 115k | Updated | Claude CodeCodex | Verified | Install |
| add-runner-evalpaperclipai/paperclipAdd or extend a Paperclip Runner protocol evaluation definition, roster, assertion, or report fixture with provenance and narrow validation. | Stars: 115k | Updated | Claude CodeCodex | Verified | Install |
| diagnose-why-work-stoppedpaperclipai/paperclipCompatibility entry point for failed, stalled, looping, or over-recovered Paperclip work. Use fix-failed-paperclip-run for the full diagnosis, regression, repair, real verification, and green-PR workflow; honor an explicitly requested diagnosis-only scope. | Stars: 115k | Updated | Claude CodeCodex | Verified | Install |
| fix-failed-paperclip-runpaperclipai/paperclipDiagnose and fix failed, stuck, slow, or incorrectly blocked Paperclip tasks and runs. Use with a task/run URL or failure report to trace the root cause, prove a regression red then green, check second-order effects, verify the real workflow, and deliver a PR with passing checks. | Stars: 115k | Updated | Claude CodeCodex | Verified | Install |