opencode-skills-collection 4.0.67 → 4.0.68

package/bundled-skills/weather-model-data-fetching/SKILL.md

@@ -0,0 +1,277 @@
+---
+name: weather-model-data-fetching
+description: "Retrieve numerical weather prediction data from public AWS S3 and HTTP archives using GRIB2 inventories, byte ranges, Herbie, provider fallbacks, and verified caching."
+category: data
+risk: safe
+source: self
+source_type: self
+date_added: "2026-09-18"
+author: ShianMike
+tags: [weather, grib2, aws-s3, herbie, noaa, nwp]
+tools: [claude, cursor, gemini, codex]
+---
+
+# Weather Model Data Fetching
+
+## Overview
+
+Fetch numerical weather prediction data without treating a multi-gigabyte GRIB2
+file as one indivisible download. Prefer an existing project adapter or Herbie;
+use direct object-store byte ranges only when the supported path cannot express
+the request.
+
+This skill covers transport, inventory selection, caching, and verification. It
+does not interpret the forecast or decide whether a model is meteorologically
+appropriate.
+
+## When to Use This Skill
+
+- A task needs GFS, GEFS, HRRR, RAP, NAM, IFS, or similar model output.
+- Data lives in a public AWS S3 bucket, NOMADS, or a public cloud mirror.
+- The input is GRIB2 and only selected variables or levels are needed.
+- A point, sounding, time series, map, or batch job needs a reliable fetch path.
+- A download is missing, partial, unexpectedly large, slow, or hard to resume.
+
+Do not activate this skill for ordinary weather-forecast questions that do not
+require model files.
+
+## Define the Request First
+
+Resolve these values before downloading:
+
+- model and product;
+- initialization cycle in UTC;
+- forecast hour and therefore valid time (`valid = initialization + lead`);
+- ensemble member when applicable;
+- variables, vertical levels, and surface fields;
+- point, region, or full-grid output;
+- cache location and when the downloaded data may be deleted.
+
+Confirm that the cycle is complete, the forecast hour exists for that cycle,
+and the requested location is inside the model domain. A recent `404` often
+means the cycle is not published yet; step back to a completed cycle instead of
+retrying indefinitely.
+
+## Choose the Smallest Retrieval Route
+
+1. Reuse the project's existing fetch/cache abstraction when it already handles
+   the model.
+2. Use Herbie for a supported GRIB2 model. It discovers AWS, NOMADS, Google,
+   Azure, and other configured sources and understands their key layouts.
+3. Use a provider-native point or Zarr endpoint when the task needs a tiny
+   spatial slice from many times or members.
+4. Use direct S3 or HTTPS object access when the key is known and no suitable
+   adapter exists.
+
+Do not recursively list a large public bucket to discover one run. Build the
+documented prefix for the model, cycle, product, forecast hour, and member, then
+probe that exact object and its inventory.
+
+Use an explicit provider priority and record the provider that succeeded. A
+fallback must refer to the same model run, product, member, and forecast hour;
+never silently substitute a different forecast.
+
+## Subset GRIB2 by Inventory
+
+GRIB2 files contain consecutive messages. A companion inventory such as
+`.idx`, `.grib2.idx`, or `.grb2.inv` records each message's starting byte.
+
+1. Fetch the small inventory first.
+2. Inspect its actual rows before writing a regex.
+3. Select exact variables, levels, and forecast-step records.
+4. Set each selected message's end byte to one less than the next message's
+   start; request the final selected message through EOF when no end is known.
+5. Coalesce adjacent selected messages into one range.
+6. Issue one `Range: bytes=START-END` request per range. S3 does not support
+   multiple ranges in one `GetObject` request.
+7. Require `206 Partial Content` and a matching `Content-Range`. If a server
+   answers `200`, do not append the whole object as though it were a fragment.
+8. Pin the object's length and identity (`ETag` and/or `Last-Modified`) while
+   downloading. Discard fragments if the object changes.
+9. Assemble into a temporary file, verify it with a GRIB decoder, then rename
+   atomically into the cache.
+
+A GRIB message contains one field over its grid. Message-range subsetting saves
+variables and levels, not geography. A point request still downloads the full
+grid for every selected message unless the provider offers a point, regional,
+Zarr, or other chunked endpoint.
+
+## Herbie Example
+
+Use the current `search` argument; `searchString` is deprecated. Start from the
+inventory, fail on an empty match, and keep the download directory explicit.
+
+```python
+from pathlib import Path
+
+from herbie import Herbie
+
+PRESSURE_FIELDS = (
+    r":(?:HGT|TMP|RH|SPFH|UGRD|VGRD):\d+(?:\.\d+)? mb:"
+)
+
+
+def fetch_hrrr_pressure_run(initialization, forecast_hour, cache_dir):
+    cache_dir = Path(cache_dir)
+    h = Herbie(
+        initialization,
+        model="hrrr",
+        product="prs",
+        fxx=forecast_hour,
+        priority=["aws", "nomads", "google", "azure"],
+        save_dir=cache_dir,
+        verbose=False,
+    )
+
+    selected = h.inventory(PRESSURE_FIELDS)
+    if selected.empty:
+        raise RuntimeError("inventory matched no pressure-level fields")
+
+    downloaded = h.download(PRESSURE_FIELDS, errors="raise")
+    path = Path(downloaded) if downloaded is not None else None
+    if path is None or not path.is_file() or path.stat().st_size == 0:
+        raise RuntimeError("GRIB2 subset was not materialized")
+
+    return path, {
+        "model": h.model,
+        "product": h.product,
+        "initialization": h.date.isoformat(),
+        "forecast_hour": h.fxx,
+        "valid_time": h.valid_date.isoformat(),
+        "provider": h.grib_source,
+        "remote_object": str(h.grib),
+        "messages": len(selected),
+    }
+```
+
+For xarray output, call `h.xarray(search, ...)` and handle either one
+`xarray.Dataset` or a list of incompatible GRIB hypercubes. Merge only groups
+whose coordinates and dimensions are compatible, and close every dataset when
+finished.
+
+## Public AWS S3 Diagnostics
+
+NOAA Open Data buckets allow unsigned reads. `--no-sign-request` prevents the
+AWS CLI from loading credentials; it does not disable TLS verification.
+
+```bash
+aws s3 ls --no-sign-request s3://noaa-hrrr-bdp-pds/hrrr.YYYYMMDD/conus/
+
+aws s3api get-object --no-sign-request \
+  --bucket noaa-hrrr-bdp-pds \
+  --key "hrrr.YYYYMMDD/conus/hrrr.tHHz.wrfprsfFF.grib2" \
+  --range "bytes=START-END" fragment.grib2
+```
+
+Use these commands to inspect a documented public object or reproduce one
+known range. For normal multi-message assembly, reuse Herbie or the project's
+tested downloader instead of scripting binary concatenation in shell.
+
+## Complete Sounding Contract
+
+A pressure-level file alone may not contain a usable surface row. Before
+building a vertical profile, require:
+
+- all published isobaric levels for geopotential height, temperature, a
+  moisture variable (dew point, relative humidity, or specific humidity), and
+  U/V wind;
+- surface pressure and terrain or surface height;
+- 2 m temperature and moisture;
+- 10 m U/V wind.
+
+Some providers split pressure and surface fields into separate products. Fetch
+and join the companion product from the same run, or reject the request with a
+list of missing fields. Do not fabricate a ground row or silently reduce the
+profile to a short mandatory-level list.
+
+After decoding, sort pressure monotonically, remove duplicate levels, normalize
+units and longitude conventions, and run the consuming project's profile QC.
+
+## Point and Batch Extraction
+
+- On one-dimensional latitude/longitude grids, labeled nearest selection may be
+  sufficient.
+- On projected or curvilinear grids with two-dimensional coordinates, use the
+  project's model-aware nearest-cell routine; verify the selected latitude,
+  longitude, and distance.
+- For many points from one model hour, fetch and decode once, then reuse it.
+- For many hours, members, or regional slices, compare the GRIB route with a
+  chunked Zarr or provider-native endpoint before scaling up.
+
+## Reliability, Cache, and Cleanup
+
+- Cache by provider, object key, object identity, and field selection. A
+  filename alone is not enough provenance.
+- Retry timeouts, `408`, `429`, and transient `5xx` responses with bounded
+  exponential backoff and jitter; honor `Retry-After`.
+- Do not retry permission errors, malformed inventories, or impossible model
+  coordinates as transient failures.
+- Bound concurrency. More range workers can increase throttling and make
+  cancellation slower.
+- Keep partial files separate from valid cache entries and resume only when the
+  remote object identity still matches.
+- For a one-shot render or export, isolate data in a request-specific temporary
+  directory and remove it in `finally` after the derived artifact is durable.
+- For an interactive viewer, retain data until the final consumer closes. Never
+  delete a shared user cache as request cleanup.
+
+Measure discovery, inventory, transfer, decode, point extraction, and rendering
+separately. A slow end-to-end request is not evidence that GRIB decoding is the
+bottleneck.
+
+## Verification Checklist
+
+- The resolved initialization time, forecast hour, valid time, product, and
+  member match the request.
+- The selected inventory is nonempty and contains every required field/level.
+- The response status, byte ranges, lengths, and object identity are consistent.
+- The final file is nonempty and opens with the intended GRIB decoder.
+- Decoded variables, units, level count, grid coordinates, and valid time are
+  plausible and explicit.
+- A point result reports the actual selected grid coordinate.
+- Cancellation leaves no file that can be mistaken for a complete cache hit.
+- Cleanup preserves the requested final artifact and removes only data owned by
+  that request.
+
+## Security & Safety Notes
+
+- Fetch only public datasets or resources the user is authorized to access.
+- Do not put cloud credentials in code, URLs, logs, examples, or skill files.
+- Keep certificate verification enabled; never solve TLS errors with
+  `--no-verify-ssl`.
+- Validate inventory-derived ranges against the remote object length before
+  allocating buffers or writing files.
+- Bound requested cycles, members, forecast hours, concurrency, disk usage, and
+  retries before a large batch.
+- Follow provider usage policies and preserve required dataset attribution.
+
+## Common Pitfalls
+
+- **No data for the newest run:** The cycle is still publishing. Use the newest
+  completed cycle and report the fallback.
+- **Subset is as large as the full file:** The inventory was missing, the regex
+  was too broad, or the server ignored `Range`.
+- **xarray returns a list:** The selected messages form multiple incompatible
+  hypercubes. Process them separately or merge only compatible groups.
+- **Point extraction is still expensive:** GRIB message ranges are not spatial
+  chunks. Use a point/regional service or Zarr when available.
+- **A cached file opens but has missing fields:** Validate the inventory contract
+  and object identity before accepting a cache hit.
+
+## Limitations
+
+- Provider key layouts, retention windows, model schedules, and Herbie templates
+  can change; verify them against current provider documentation.
+- Variable subsetting requires a usable remote inventory. Without one, download
+  the full object or use a different provider.
+- This skill does not validate forecast skill, scientific suitability, or
+  proprietary-provider credentials and quotas.
+
+## Additional Resources
+
+- [Herbie documentation](https://herbie.readthedocs.io/)
+- [Herbie source and model templates](https://github.com/blaylockbk/Herbie)
+- [NOAA fast GRIB2 downloads with inventories](https://nomads.ncep.noaa.gov/info.php?page=fastdownload)
+- [Amazon S3 `GetObject` byte ranges](https://docs.aws.amazon.com/AmazonS3/latest/API/API_GetObject.html)
+- [NOAA HRRR on the AWS Registry of Open Data](https://registry.opendata.aws/noaa-hrrr-pds/)
+- [NOAA GFS on the AWS Registry of Open Data](https://registry.opendata.aws/noaa-gfs-bdp-pds/)

package/bundled-skills/weather-observation-fetching/SKILL.md

@@ -0,0 +1,246 @@
+---
+name: weather-observation-fetching
+description: "Retrieve surface and upper-air weather observations from authoritative APIs and archives with station identity, time, units, and quality flags preserved."
+category: data
+risk: safe
+source: self
+source_type: self
+date_added: "2026-09-19"
+author: ShianMike
+tags: [weather, observations, metar, radiosonde, noaa, quality-control]
+tools: [claude, cursor, gemini, codex]
+---
+
+# Weather Observation Fetching
+
+## Overview
+
+Retrieve measured surface and upper-air weather reports without losing station
+identity, observation time, units, raw values, or provider quality flags. Pick
+the source by observation type and retention need, then validate the returned
+records before normalization.
+
+This skill covers METARs, historical surface observations, radiosondes, and
+station metadata. It excludes model output, radar volumes, and satellite
+imagery.
+
+## When to Use This Skill
+
+- A task needs recent METAR observations for named stations or a small region.
+- Historical hourly or synoptic surface data is needed from NOAA NCEI.
+- A sounding workflow needs observed radiosonde profiles rather than model
+  profiles.
+- Station identifiers, relocations, instruments, or metadata must be resolved.
+- A fetch returned duplicate, stale, unit-ambiguous, or quality-flagged values.
+
+Do not use forecast products as observations, and do not substitute a nearby
+model grid point for a missing station report without explicit approval.
+
+## Choose the Source
+
+| Need | Preferred source | Notes |
+| --- | --- | --- |
+| Recent aviation surface reports | NOAA Aviation Weather Center Data API | Query a small station/time set; use published cache files for bulk current data. |
+| Historical global surface reports | NOAA NCEI Integrated Surface Database (ISD) | Preserve USAF/WBAN identity, units, and QC fields. |
+| Historical or recent radiosondes | NOAA NCEI IGRA | Use the station inventory and retain level and QC metadata. |
+| Station history and identifier changes | NOAA NCEI station history/HOMR | Resolve moves, renames, and observing-platform changes. |
+
+Prefer an existing project adapter when it already handles the provider's
+schema, retries, and cache. Record the exact endpoint or archive object used.
+
+## Define the Observation Request
+
+Resolve these values before fetching:
+
+- observation type and variables;
+- station identifier system, not just the identifier string;
+- start and end instants in UTC, including interval inclusivity;
+- maximum acceptable observation age;
+- raw, decoded, or both output forms;
+- required quality flags and policy for rejected values;
+- output units and missing-value representation;
+- cache location and retention.
+
+For spatial queries, also define the search geometry, distance limit, and how a
+station is selected. Return the selected station and distance rather than
+silently using the nearest report.
+
+## Fetch Recent METARs
+
+The Aviation Weather Center exposes machine-readable METAR data under
+`/api/data/metar`. Send a descriptive user agent, keep the query narrow, and
+handle a valid `204 No Content` separately from an error.
+
+```python
+import json
+from urllib.error import HTTPError
+from urllib.parse import urlencode
+from urllib.request import Request, urlopen
+
+
+def fetch_metars(stations, hours=2):
+    station_ids = sorted({station.strip().upper() for station in stations})
+    if not station_ids or any(len(station) != 4 for station in station_ids):
+        raise ValueError("use one or more four-character ICAO station IDs")
+    if not 1 <= hours <= 24:
+        raise ValueError("hours must be between 1 and 24 for this narrow query")
+
+    query = urlencode({
+        "ids": ",".join(station_ids),
+        "format": "json",
+        "hours": hours,
+    })
+    request = Request(
+        f"https://aviationweather.gov/api/data/metar?{query}",
+        headers={"User-Agent": "weather-observation-fetching/1.0 contact@example.org"},
+    )
+    try:
+        with urlopen(request, timeout=30) as response:
+            if response.status == 204:
+                return []
+            records = json.load(response)
+    except HTTPError as exc:
+        if exc.code == 429:
+            raise RuntimeError("AWC rate limit reached; honor Retry-After") from exc
+        raise
+
+    if not isinstance(records, list):
+        raise RuntimeError("unexpected METAR response shape")
+    return records
+```
+
+Replace the example contact address with an appropriate project contact. For a
+large current snapshot, download the provider's compressed cache file once
+instead of issuing many station queries.
+
+## Fetch Historical Surface Data
+
+For ISD:
+
+1. Resolve the station using the current station inventory and its USAF/WBAN
+   identifiers.
+2. Confirm that the station's coverage overlaps the requested time range.
+3. Use bulk HTTPS files for a large historical request; avoid one network call
+   per observation.
+4. Preserve the original report and source/QC codes before converting units.
+5. Treat trace values, missing sentinels, and calm or variable winds according
+   to the data format documentation.
+6. Join station metadata by both identifier and effective date when station
+   history matters.
+
+Do not assume one station identifier always represents an unchanged location or
+instrument throughout its archive.
+
+## Fetch Radiosonde Profiles
+
+For IGRA:
+
+1. Search the station inventory by identifier or location and verify the
+   station's record period.
+2. Fetch the station file covering the requested dates rather than scraping an
+   interactive page.
+3. Select by the report's UTC time and retain nominal, launch, and release times
+   when the source supplies them.
+4. Preserve pressure, height, temperature, moisture, wind, level type, and QC
+   fields. Standard and significant levels are both scientifically relevant.
+5. Sort the profile only after parsing; do not invent levels or interpolate
+   across large gaps during acquisition.
+6. Report an absent launch or incomplete profile explicitly.
+
+Many upper-air stations usually report near 00 and 12 UTC, but the archive is
+the authority. Do not manufacture a schedule or select a different day solely
+because a nominal time is missing.
+
+## Normalize Without Erasing Provenance
+
+Each normalized record should retain:
+
+- provider and dataset;
+- station identifier plus identifier scheme;
+- station latitude, longitude, elevation, and metadata effective date;
+- observation time in UTC and, when available, receipt or ingestion time;
+- raw report or raw archive row;
+- decoded values with explicit units;
+- provider quality flags and local QC decisions;
+- retrieval time, source URL/object, and response identity.
+
+Store original and converted values side by side when a conversion could affect
+rounding. Never use the HTTP `Last-Modified` timestamp as the observation time.
+
+## Quality Control and Deduplication
+
+- Treat provider flags as data, not decoration. Define which flags are accepted,
+  rejected, or retained with warnings.
+- Deduplicate on provider identity, station, observation time, and report type.
+  When corrected reports exist, preserve the correction lineage.
+- Check physical ranges only after handling missing and trace encodings.
+- Verify wind direction conventions, temperature scales, pressure units, and
+  precipitation accumulation periods before combining sources.
+- Keep station time, observation time, and ingestion time distinct.
+- Flag stale reports against the request's maximum age rather than returning
+  them as current conditions.
+
+## Reliability and Caching
+
+- Honor provider request limits, `Retry-After`, and published bulk-download
+  guidance.
+- Retry timeouts, `408`, `429`, and transient `5xx` failures with bounded
+  backoff and jitter.
+- Cache immutable archive files by URL/object identity and current API responses
+  for no longer than their update cadence permits.
+- Write downloads to a temporary path, validate content and expected date range,
+  then rename atomically.
+- Keep partial files separate from accepted cache entries.
+- For one-shot processing, remove request-owned temporary observations in
+  `finally` only after the derived artifact and provenance record are durable.
+
+## Verification Checklist
+
+- The station identifier scheme and station metadata are explicit.
+- All selected observations fall inside the requested UTC interval.
+- The report time, receipt time, and retrieval time are not conflated.
+- Units, missing sentinels, trace values, and QC flags are handled explicitly.
+- Raw reports or rows remain available for audit.
+- Duplicate and corrected reports follow a documented rule.
+- A no-data response is distinguished from provider failure.
+- The final result reports stale, incomplete, or rejected observations.
+
+## Security & Safety Notes
+
+- Use only public endpoints or data the user is authorized to access.
+- Do not place API keys, credentials, signed URLs, or private station data in
+  examples, logs, caches, or provenance manifests.
+- Keep TLS certificate verification enabled.
+- Encode query parameters rather than concatenating untrusted station input into
+  a URL.
+- Bound station count, time span, response size, retries, and parallelism.
+- Follow provider terms, rate limits, and attribution requirements.
+
+## Common Pitfalls
+
+- **The latest METAR is old:** The station has not reported recently. Apply the
+  maximum-age contract and report staleness.
+- **A station lookup returns the wrong site:** ICAO, WMO, USAF/WBAN, and IGRA
+  identifiers were treated as interchangeable. Preserve the identifier scheme.
+- **Temperatures look extreme:** Missing sentinels or units were converted as
+  real values. Parse format metadata before unit conversion.
+- **A sounding has too few levels:** Only mandatory levels were retained or the
+  launch was incomplete. Preserve significant levels and surface data.
+- **An archive record moved:** Station history changed. Join metadata by its
+  effective period and record the selected version.
+
+## Limitations
+
+- Provider schemas, retention windows, station inventories, and usage limits can
+  change; consult current official documentation.
+- Quality flags identify known conditions but do not guarantee that a
+  measurement is scientifically suitable for a particular analysis.
+- This skill does not perform radar retrieval, satellite retrieval, model-data
+  fetching, or forecast verification.
+
+## Additional Resources
+
+- [NOAA Aviation Weather Center Data API](https://aviationweather.gov/data/api/)
+- [NOAA NCEI Integrated Surface Database](https://www.ncei.noaa.gov/products/land-based-station/integrated-surface-database)
+- [NOAA NCEI Integrated Global Radiosonde Archive](https://www.ncei.noaa.gov/products/weather-balloon/integrated-global-radiosonde-archive)
+- [NOAA NCEI station histories](https://www.ncei.noaa.gov/products/land-based-station/station-histories)

package/bundled-skills/wiki-yylo/SKILL.md

@@ -0,0 +1,114 @@
+---
+name: wiki-yylo
+description: Use YYLO Ledger wiki Records as durable project knowledge. Search before
+  creating, classify information correctly, and make revision-safe Markdown updates
+  without editing Ledger storage directly.
+category: project-management
+risk: safe
+source: https://github.com/yylo-dev/yylo-skills
+source_repo: yylo-dev/yylo-skills
+source_type: community
+date_added: '2026-09-19'
+license: MIT
+license_source: https://github.com/yylo-dev/yylo-skills/blob/main/LICENSE
+compatibility: Requires the `yy` CLI with the `wiki` record group installed (controller
+  or standalone). Revision-safe Markdown updates only; never edit Ledger storage directly.
+argument-hint: '[wiki question or knowledge to find/create/update]'
+enable-shell-directives: true
+---
+
+# Use YYLO wiki Records
+
+Treat Ledger as the source of truth. Use `yy ledger` in a YYLO controller and
+`yylo-ledger` in a standalone Ledger project. Inspect `COMMAND wiki --help` before
+acting; if `wiki` is absent, the installed Ledger version does not expose native
+Record commands and must not be bypassed with direct file edits.
+
+## Choose the right record
+
+Before writing, decide where the information belongs:
+
+- **Wiki**: durable explanatory project or domain knowledge that future work must discover.
+- **Task**: scoped requested work, status, dependencies, acceptance criteria, and completion evidence.
+- **Workflow**: validated structured steps, not prose guidance.
+- **Artifact**: generated evidence, reports, receipts, logs, model output, or binary payloads.
+- **Source documentation**: documentation released and versioned with product code.
+
+Do not put secrets, caches, session transcripts, temporary status, bulky generated
+evidence, or owner-only operational receipts in a wiki.
+
+## Discover before creating
+
+Use bounded summary searches first. Resolve records by immutable ID whenever one
+is known; slugs and aliases are discovery conveniences, not replacement identity.
+
+```bash
+yy ledger wiki search --text "deployment policy" --projection summary --limit 20 -f json
+yy ledger wiki get RECORD_ID -f json
+yy ledger wiki get RECORD_ID --raw
+```
+
+Use `--scope archive|all` only when the request requires cold records. Request
+`full` projection only when payload bytes are necessary.
+
+## Create durable Markdown
+
+Use a file or stdin for substantial or shell-sensitive content:
+
+```bash
+yy ledger wiki create --title "Service ownership" --file ownership.md
+yy ledger wiki create --title "Incident notes" --file - < incident-notes.md
+```
+
+Choose a stable title, namespace, slug, aliases, and relations deliberately. Keep
+one topic per Record and link related immutable Record IDs rather than duplicating
+truth.
+
+## Update safely
+
+1. Read the current Record and revision.
+2. Preserve its immutable ID and inspect history when intent is unclear.
+3. Follow `wiki update --help` for the installed compare-and-replace controls.
+4. Supply the expected revision and required preimage/digest evidence.
+5. Use file transport; do not rewrite Ledger files yourself.
+6. Read back the resulting revision and receipt.
+
+A revision or preimage mismatch means the source changed: reread and reconcile.
+Never force past concurrent edits. Archive is a lifecycle transition, not delete.
+
+Use `--front-matter` only for canonical front-matter interchange and `--rendered`
+for inert, HTML-escaped rendering. Use `history RECORD_ID` to understand revisions;
+do not infer history from the latest payload alone.
+
+## Project wiki boundary
+
+Portable controller guidance may live under the controller wiki, while project
+and domain pages retain project-owned paths. Package-managed and project-owned
+pages can coexist. Migration, runtime replacement, exceptional merge recovery,
+release, deployment, and cold-archive maintenance remain authoritative runbooks,
+not content to summarize into an everyday global skill.
+
+## Complete request
+
+$ARGUMENTS
+
+## When to Use
+
+- You need durable project or domain knowledge as YYLO Ledger wiki Records (search, create, revision-safe update).
+- You must first decide the record belongs in the wiki rather than a task, workflow, artifact, or product docs.
+
+## Limitations
+
+- Search before creating; one topic per Record, linked by immutable Record IDs.
+- Never store secrets, caches, session transcripts, temporary status, bulky generated evidence, or owner-only receipts in a wiki.
+- Updates bind to the expected revision/preimage; on mismatch reread and reconcile - never force past concurrent edits. Archive is a lifecycle transition, not deletion.
+- If the `wiki` group is absent, fail closed and request a Ledger upgrade.
+
+### Example
+
+```bash
+yy ledger wiki search --text "deployment policy" --projection summary --limit 20 -f json
+yy ledger wiki get RECORD_ID -f json
+```
+
+> Adapted from [yylo-dev/yylo-skills](https://github.com/yylo-dev/yylo-skills) (MIT) - v2.0.1; frontmatter, When to Use/Limitations, and safety boundaries added for upstream compliance.