[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6MA772CMVPYBDCBD03PFZC8":3,"skill-same:01M6MA772CMVPYBDCBD03PFZC8":136},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":40,"linkOnly":41,"markdownContent":42,"markdownBody":43,"frontMatter":44,"refName":52,"commitSha":53,"skillDirectory":54,"forksCount":55,"repositoryHtmlUrl":56,"repositoryHomepageUrl":57,"ownerHtmlUrl":58,"ownerBlogUrl":59,"ownerDescription":60,"updatedAt":61,"install":62,"resources":75},"01M6MA772CMVPYBDCBD03PFZC8","gtars",null,"Inspects and plans work with Gtars, the Rust\u002FPython\u002FCLI toolkit for genomic intervals: BED RegionSet set algebra (reduce, setdiff, intersect, closest, cluster, gaps), overlap counts, consensus peak sets, coverage, tokenizers, fragment processing, and refget\u002FBEDbase stores. Use when merging or comparing BED interval sets, building a consensus or universe from peaks, tokenizing regions for ML models, working with fragment files, or planning refget sequence-collection and BEDbase cache use. Use when checking BED coordinates, contigs, or strand before running gtars. Not for general BAM\u002FVCF processing; use samtools or bedtools for that.","hashgraph-online\u002Fawesome-codex-plugins","hashgraph-online","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F160371962?v=4","01M6CNB6NRJVNNGN14TR01Q6D3","HOL (Hashgraph Online)","01M6JT1HSKKG9V68KQAQGBA7XP",1575,"MIT","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002FKalarisLabs\u002Fresearch-agent-skills\u002Fskills\u002Fgtars\u002FSKILL.md",[],[5,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39],"files","when","2024","reduce","remote","processing","vcf","bed","free","local","caching","cli","only","coverage","consensus","work","Bash","collection","SET","Meta","rust","2026-11-10 10:12:30",false,"---\nname: gtars\ndescription: 'Inspects and plans work with Gtars, the Rust\u002FPython\u002FCLI toolkit for genomic intervals: BED RegionSet set algebra (reduce, setdiff, intersect, closest, cluster, gaps), overlap counts, consensus peak sets, coverage, tokenizers, fragment processing, and refget\u002FBEDbase stores. Use when merging or comparing BED interval sets, building a consensus or universe from peaks, tokenizing regions for ML models, working with fragment files, or planning refget sequence-collection and BEDbase cache use. Use when checking BED coordinates, contigs, or strand before running gtars. Not for general BAM\u002FVCF processing; use samtools or bedtools for that.'\nlicense: MIT\ncompatibility: Python bindings require Python 3.10+ and gtars 0.9.2. The Rust meta-crate and gtars-cli are 0.9.0 and require a Rust toolchain supporting Edition 2024; upstream declares no rust-version. Bundled audit CLIs use only Python 3.10+ standard library and are local\u002Fnetwork-free. Remote constructors, pretrained tokenizers, refget, and BEDbase caching require explicit network and storage approval.\nallowed-tools: Read Write Edit Bash Glob\nmetadata:\n  version: '1.3'\n  category: life-sciences\n  maintainer: Kalaris Labs\n  tags: genomic intervals, BED files, peak sets, genomics, bioinformatics\n---\n\n# Gtars\n\nGtars provides native Rust implementations, Python bindings, and a feature-gated\n`gtars` binary for genomic interval and reference-sequence work. Start with the\nbundled local inspectors; call upstream code only after the data contract,\nprovenance, resource bounds, and side effects are explicit.\n\n## Verified snapshot (2026-07-23)\n\n- Python: [`gtars==0.9.2`](https:\u002F\u002Fpypi.org\u002Fproject\u002Fgtars\u002F), released\n  2026-06-17, `Requires-Python >=3.10`.\n- Rust meta-crate: [`gtars=0.9.0`](https:\u002F\u002Fcrates.io\u002Fcrates\u002Fgtars), released\n  2026-06-15. Its default feature set is empty.\n- CLI crate\u002Fbinary: [`gtars-cli=0.9.0`](https:\u002F\u002Fcrates.io\u002Fcrates\u002Fgtars-cli);\n  the installed binary is named `gtars`.\n- Direct refget crate: [`gtars-refget=0.9.1`](https:\u002F\u002Fcrates.io\u002Fcrates\u002Fgtars-refget),\n  released 2026-06-17. `gtars=0.9.0` itself pins its component release set, which\n  includes refget 0.9.0.\n- Upstream intentionally versions workspace crates, Python bindings, and CLI\n  independently. Do not assume matching numbers mean matching artifacts.\n- The published docs changelog stops at 0.5.1. API examples here were checked\n  against the 0.9.2 Python stubs\u002Fruntime and the `v0.9.0` CLI\u002FRust source.\n\nThe `license: MIT` field covers this skill. Published `gtars` crates declare MIT,\nwhile the GitHub repository currently displays BSD-2-Clause at the root; verify\nthe exact artifact's license before redistribution.\n\n## Native-code trust gate and exact pins\n\nThe Python wheel contains a PyO3 native extension. Cargo installation compiles a\nnative binary and can run dependency build scripts. Treat either path as code\nexecution:\n\n1. Confirm the official PyPI\u002Fcrates.io\u002FGitHub owner and immutable version.\n2. Review filenames, platform tags, release provenance, license, and SHA-256.\n   GitHub's v0.9.0 binary release includes per-archive `.sha256` sidecars.\n3. Never run an untrusted prebuilt binary, wheel, source tree, Cargo build script,\n   or archive installer. Use isolation and CPU\u002FRAM\u002Fdisk\u002Ftime limits.\n4. Keep a lockfile and artifact hashes with the analysis manifest.\n\nAfter that review, create an isolated Python environment:\n\n```bash\nuv venv --python 3.11 .venv-gtars\nuv pip install --dry-run --python .venv-gtars\u002Fbin\u002Fpython \"gtars==0.9.2\"\nuv pip install --python .venv-gtars\u002Fbin\u002Fpython \"gtars==0.9.2\"\n.venv-gtars\u002Fbin\u002Fpython -c \\\n  \"import gtars; assert gtars.__version__ == '0.9.2'; print(gtars.__version__)\"\n```\n\nFor the reviewed CLI source release:\n\n```bash\ncargo install gtars-cli --version 0.9.0 --locked\ngtars --version\ngtars --help\n```\n\nFor a Rust project, pin the wrapper exactly and enable only required features:\n\n```toml\n[dependencies]\ngtars = { version = \"=0.9.0\", default-features = false, features = [\n  \"core\", \"overlaprs\", \"uniwig\", \"tokenizers\", \"refget\"\n] }\n```\n\nUse `gtars-refget = \"=0.9.1\"` directly only when the newer direct component API is\nrequired and compatibility has been tested. Do not replace these pins with a Git\nbranch or an unreviewed release.\n\n## Genomic data contract\n\nApply this contract before every operation:\n\n1. **Coordinates:** BED intervals are 0-based and half-open: `[start, end)`.\n   Require `0 \u003C= start \u003C end \u003C= contig_length`. Gtars coordinates are `u32`, so\n   reject values above `4,294,967,295`.\n2. **Assembly:** record an assembly accession\u002Fversion and the SHA-256 of the exact\n   chromosome-sizes or refget sequence-collection metadata. Never infer assembly\n   from filenames or `chr` prefixes.\n3. **Contigs:** compare names exactly. `1` and `chr1`, alternate loci, decoys, and\n   mitochondrial aliases are not interchangeable. Rename or liftover only as a\n   separately reviewed transformation.\n4. **Sorting:** preserve the original file, then sort a copy by chromosome-sizes\n   order and numeric start\u002Fend when the operation requires it. Python\n   `RegionSet(path)` currently sorts lexicographically by contig and start while\n   loading; do not rely on original row order afterward.\n5. **Strand:** BED6 uses `+`, `-`, or `.`. `Region.rest` retains trailing BED\n   fields, but a file-backed Python `RegionSet` currently initializes its separate\n   `strands` vector to `*`. Several set operations drop strand. Preserve and\n   validate strand externally when it is scientifically meaningful.\n6. **Duplicates\u002Fadjacency:** choose policies explicitly. `reduce()` and consensus\n   merge overlapping **and adjacent** intervals; ordinary half-open overlap does\n   not treat `[0,10)` and `[10,20)` as overlapping.\n\nRun the local validator first:\n\n```bash\npython3 -B scripts\u002Fbed_validator.py \\\n  --input data.bed.gz \\\n  --assembly GRCh38.p14 \\\n  --chrom-sizes GRCh38.p14.chrom.sizes \\\n  --require-sorted\n```\n\n## Safe local workflow\n\n1. Inventory local files, checksums, assembly, contig dictionary, coordinate\n   system, strand policy, patient\u002Freplicate groups, and intended outputs.\n2. Validate BED\u002Ffragments and estimate work. Pilot a small synthetic file.\n3. Choose Python, CLI, or Rust from the documented surface; do not translate API\n   names by guesswork.\n4. Set hard limits for input bytes\u002Frecords\u002Ffiles, threads\u002Fjobs, memory, temporary\n   disk, output size, and wall time.\n5. Run in a dedicated output directory. Refuse collisions unless overwrite was\n   explicitly approved.\n6. Revalidate output sorting, bounds, row counts, checksums, and provenance.\n\n## Current Python core\n\nImports are from submodules, not the `gtars` top level:\n\n```python\nfrom gtars.models import Region, RegionSet\n\nquery = RegionSet.from_regions(\n    [\n        Region(chr=\"chr1\", start=100, end=200, rest=None),\n        Region(chr=\"chr1\", start=300, end=400, rest=None),\n    ],\n    strands=[\"+\", \"-\"],\n)\nuniverse = RegionSet.from_vectors(\n    [\"chr1\", \"chr1\"],\n    [150, 500],\n    [350, 600],\n)\n\ncounts = query.count_overlaps(universe)       # one count per query region\nflags = query.any_overlaps(universe)          # one bool per query region\nindices = query.find_overlaps(universe)       # indices into universe\npieces = query.intersect_all(universe)        # all intersection fragments\nfraction = query.coverage(universe)           # fraction of query bp covered\n```\n\n`RegionSet.sort()` mutates and returns `None`. Set algebra includes `reduce`,\n`setdiff`, `pintersect` (pairs by index), `concat`, `union`, `jaccard`,\n`coverage`, `overlap_coefficient`, `intersect_all`, `closest`, `cluster`, and\n`gaps`. Read `references\u002Fpython-api.md` before relying on ordering or strand.\n\nConsensus is a Python binding in a different module:\n\n```python\nfrom gtars.genomic_distributions import consensus\n\nrows = consensus([query, universe])\n# rows: [{\"chr\": ..., \"start\": ..., \"end\": ..., \"count\": ...}, ...]\n```\n\nSignal-track generation is **not** exposed as `gtars.uniwig` in Python 0.9.2;\nuse the reviewed CLI or Rust API. `RegionSet.coverage()` is a base-pair set metric,\nnot a WIG\u002FbigWig generator.\n\n## Tokenizers, fragments, and reference stores\n\nUse only local constructors by default:\n\n```python\nfrom gtars.models import RegionSet\nfrom gtars.tokenizers import Tokenizer\n\ntokenizer = Tokenizer.from_bed(\"reviewed-universe.bed\")\nregions = RegionSet(\"local-query.bed\")\ntokens = tokenizer.tokenize(regions)\nencoding = tokenizer(regions)\nids = encoding[\"input_ids\"]\n```\n\n`Tokenizer.from_pretrained(name)` contacts Hugging Face and writes its cache when\nthe argument is not an existing local directory; it exposes no revision or cache\nargument. Obtain explicit approval, fetch an immutable revision through a reviewed\nmechanism, verify checksums, then pass the local snapshot directory. See\n`references\u002Ftokenizers.md`.\n\nFor refget, prefer `RefgetStore.in_memory()` or `RefgetStore.open_local(path)`.\n`open_remote(cache_path, remote_url)` contacts a remote service, creates\u002Fuses a\nlocal cache, and performs on-demand range reads. See `references\u002Frefget.md`.\n\n## Network and cache gate\n\nNo download or cache write is implicit in this skill. Before any network-capable\nupstream call:\n\n- obtain explicit user approval for the exact host, endpoint, data, and cache;\n- allowlist HTTPS hosts and reject unreviewed redirects;\n- record immutable revision\u002Fidentifier, retrieval time, expected SHA-256 and\n  domain digest, assembly accession, size quota, and provenance;\n- disclose sensitive BED coordinates, barcodes, sample labels, and reference\n  choices that could leave the approved environment;\n- validate downloaded content as untrusted before using it.\n\nImportant side effects:\n\n- `RegionSet(path)` has HTTP support; a nonexistent local string may be treated as\n  a URL. Check that the local path exists before construction.\n- `Tokenizer.from_pretrained` may download `universe.bed.gz` into the Hugging Face\n  cache.\n- `RefgetStore.on_disk` creates\u002Fwrites a store. `open_remote` loads remote metadata\n  and enables persistence by default.\n- `gtars bbcache` creates cache directories even when constructing the client.\n  Cache\u002Fdownload commands use `BBCLIENT_CACHE` (default `~\u002F.bbcache`) and\n  `BEDBASE_API` (default `https:\u002F\u002Fapi.bedbase.org`).\n\n## Sensitive metadata and leakage\n\nGenomic intervals, rare loci, barcodes, sample names, phenotypes, and assembly\nchoices can be identifying. Keep full paths and raw coordinates out of logs;\ndefault bundled reports redact paths and emit only counts\u002Fchecksums.\n\nFreeze splits by patient\u002Fdonor first, then keep all technical and biological\nreplicates in the same split. Fit consensus sets, universes, tokenizers, scaling,\nthresholds, and QC rules on training data only. Do not create a universe from all\nsamples and then split: that leaks validation\u002Ftest locus support. Record excluded\nsamples and replicate aggregation separately.\n\n## Bundled deterministic CLIs\n\nAll six helpers reject URLs, traversal, symlinks, and special files; apply byte,\nrecord, file, coordinate, and worker caps; use no network or gtars import; and\nwrite no output files. Plans contain fixed argv templates and never launch them.\n\n```bash\npython3 -B scripts\u002Fbed_validator.py --help\npython3 -B scripts\u002Fexecution_plan.py --help\npython3 -B scripts\u002Ftokenizer_manifest.py --help\npython3 -B scripts\u002Frefget_digest_plan.py --help\npython3 -B scripts\u002Fcoverage_preflight.py --help\npython3 -B scripts\u002Fartifact_inspector.py --help\n```\n\nRun synthetic tests without bytecode:\n\n```bash\nPYTHONDONTWRITEBYTECODE=1 python3 -B -m unittest discover \\\n  -s tests\u002Fgtars -p 'test_*.py' -v\n```\n\n## Migration traps removed in 1.1\n\nDo not use stale examples containing `gtars.RegionSet`,\n`RegionSet.from_bed`, `TreeTokenizer`, `gtars.igd.build_index`,\n`gtars.uniwig.coverage_from_bed`, `gtars.RefgetStore`, global\n`set_option`\u002F`set_log_level`, `parallel_apply`, or invented exception classes.\nCLI forms such as `uniwig generate`, `igd build`, `scoring score`, and\n`fragsplit cluster-split` are also stale for 0.9.0.\n\nUpstream's published docs and stubs have some drift (for example the older\n`GlobalRefgetStore` tutorial and incomplete 0.9.2 stubs). Prefer installed\nsignature smoke tests plus immutable tagged source when they conflict.\n\n## Bundled references\n\nThese are the only six bundled references; all links are local and present:\n\n- `references\u002Fpython-api.md` — exact Python 0.9.2 imports and behavior\n- `references\u002Foverlap.md` — overlap\u002Fcount\u002Fset algebra and consensus semantics\n- `references\u002Fcoverage.md` — uniwig, bigWig, coverage, sorting, and resources\n- `references\u002Ftokenizers.md` — tokenizer\u002Funiverse and fragment compatibility\n- `references\u002Frefget.md` — digests, stores, BEDbase, network\u002Fcache controls\n- `references\u002Fcli.md` — CLI 0.9.0 commands, features, and migrations\n\n## Agent operating procedure\n\n1. **Check the environment.** Confirm tool versions, the reference genome\u002Fannotation build and the input formats (FASTQ, BAM, VCF, h5ad).\n2. **Pin down the inputs.** Confirm formats, identifiers and parameters from the data or the user. Ask rather than guess any value that changes the result.\n3. **Run a small version first.** Run the pipeline on a small subset (one sample, one chromosome, a few thousand cells) first.\n4. **Execute the full task** using the instructions and references above.\n5. **Validate the result.** Check QC metrics, sample identities, genome build consistency and batch effects before interpreting results.\n6. **Report.** State what was run (versions, commands, parameters), what was checked, and what is still uncertain.\n\n| If this happens | Do this |\n|---|---|\n| Genome builds or identifiers do not match between inputs | Stop and harmonize (liftover, ID mapping) before continuing. |\n| A function, flag or endpoint in these instructions is missing in the installed version | Check the installed version's own documentation (`help()`, `--help`, official docs), adapt, and tell the user. Never invent an API. |\n| A required input, identifier or parameter is ambiguous | Ask the user, or state the assumption explicitly before running. |\n\n**Integrity rules**\n\n- Never fabricate results, parameters, identifiers, citations or statistics. If something cannot be run or verified, say so plainly.\n- Do not interpret biological significance beyond what the statistics support; report multiple-testing correction.\n- Treat version-specific details here as possibly outdated: confirm them against the official documentation for the installed version.\n- Ask before actions that cost money, consume shared GPUs or cloud quota, touch personal or patient data, or cannot be undone.\n\n## Related skills\n\n- `geniml`: Plans and audits local genomic-interval machine learning workflows with Geniml (0.8.4) and Gtars: validates BED files against chromosome si…\n- `polars-bio`: Python library polars-bio for genomic interval operations and bioinformatics file I\u002FO on Polars DataFrames, built on Arrow and DataFusion.\n- `genomic-coordinates`: Convert genomic intervals between coordinate conventions, normalise and compare variant representations, and detect assembly or contig-nami…\n","# Gtars\n\nGtars provides native Rust implementations, Python bindings, and a feature-gated\n`gtars` binary for genomic interval and reference-sequence work. Start with the\nbundled local inspectors; call upstream code only after the data contract,\nprovenance, resource bounds, and side effects are explicit.\n\n## Verified snapshot (2026-07-23)\n\n- Python: [`gtars==0.9.2`](https:\u002F\u002Fpypi.org\u002Fproject\u002Fgtars\u002F), released\n  2026-06-17, `Requires-Python >=3.10`.\n- Rust meta-crate: [`gtars=0.9.0`](https:\u002F\u002Fcrates.io\u002Fcrates\u002Fgtars), released\n  2026-06-15. Its default feature set is empty.\n- CLI crate\u002Fbinary: [`gtars-cli=0.9.0`](https:\u002F\u002Fcrates.io\u002Fcrates\u002Fgtars-cli);\n  the installed binary is named `gtars`.\n- Direct refget crate: [`gtars-refget=0.9.1`](https:\u002F\u002Fcrates.io\u002Fcrates\u002Fgtars-refget),\n  released 2026-06-17. `gtars=0.9.0` itself pins its component release set, which\n  includes refget 0.9.0.\n- Upstream intentionally versions workspace crates, Python bindings, and CLI\n  independently. Do not assume matching numbers mean matching artifacts.\n- The published docs changelog stops at 0.5.1. API examples here were checked\n  against the 0.9.2 Python stubs\u002Fruntime and the `v0.9.0` CLI\u002FRust source.\n\nThe `license: MIT` field covers this skill. Published `gtars` crates declare MIT,\nwhile the GitHub repository currently displays BSD-2-Clause at the root; verify\nthe exact artifact's license before redistribution.\n\n## Native-code trust gate and exact pins\n\nThe Python wheel contains a PyO3 native extension. Cargo installation compiles a\nnative binary and can run dependency build scripts. Treat either path as code\nexecution:\n\n1. Confirm the official PyPI\u002Fcrates.io\u002FGitHub owner and immutable version.\n2. Review filenames, platform tags, release provenance, license, and SHA-256.\n   GitHub's v0.9.0 binary release includes per-archive `.sha256` sidecars.\n3. Never run an untrusted prebuilt binary, wheel, source tree, Cargo build script,\n   or archive installer. Use isolation and CPU\u002FRAM\u002Fdisk\u002Ftime limits.\n4. Keep a lockfile and artifact hashes with the analysis manifest.\n\nAfter that review, create an isolated Python environment:\n\n```bash\nuv venv --python 3.11 .venv-gtars\nuv pip install --dry-run --python .venv-gtars\u002Fbin\u002Fpython \"gtars==0.9.2\"\nuv pip install --python .venv-gtars\u002Fbin\u002Fpython \"gtars==0.9.2\"\n.venv-gtars\u002Fbin\u002Fpython -c \\\n  \"import gtars; assert gtars.__version__ == '0.9.2'; print(gtars.__version__)\"\n```\n\nFor the reviewed CLI source release:\n\n```bash\ncargo install gtars-cli --version 0.9.0 --locked\ngtars --version\ngtars --help\n```\n\nFor a Rust project, pin the wrapper exactly and enable only required features:\n\n```toml\n[dependencies]\ngtars = { version = \"=0.9.0\", default-features = false, features = [\n  \"core\", \"overlaprs\", \"uniwig\", \"tokenizers\", \"refget\"\n] }\n```\n\nUse `gtars-refget = \"=0.9.1\"` directly only when the newer direct component API is\nrequired and compatibility has been tested. Do not replace these pins with a Git\nbranch or an unreviewed release.\n\n## Genomic data contract\n\nApply this contract before every operation:\n\n1. **Coordinates:** BED intervals are 0-based and half-open: `[start, end)`.\n   Require `0 \u003C= start \u003C end \u003C= contig_length`. Gtars coordinates are `u32`, so\n   reject values above `4,294,967,295`.\n2. **Assembly:** record an assembly accession\u002Fversion and the SHA-256 of the exact\n   chromosome-sizes or refget sequence-collection metadata. Never infer assembly\n   from filenames or `chr` prefixes.\n3. **Contigs:** compare names exactly. `1` and `chr1`, alternate loci, decoys, and\n   mitochondrial aliases are not interchangeable. Rename or liftover only as a\n   separately reviewed transformation.\n4. **Sorting:** preserve the original file, then sort a copy by chromosome-sizes\n   order and numeric start\u002Fend when the operation requires it. Python\n   `RegionSet(path)` currently sorts lexicographically by contig and start while\n   loading; do not rely on original row order afterward.\n5. **Strand:** BED6 uses `+`, `-`, or `.`. `Region.rest` retains trailing BED\n   fields, but a file-backed Python `RegionSet` currently initializes its separate\n   `strands` vector to `*`. Several set operations drop strand. Preserve and\n   validate strand externally when it is scientifically meaningful.\n6. **Duplicates\u002Fadjacency:** choose policies explicitly. `reduce()` and consensus\n   merge overlapping **and adjacent** intervals; ordinary half-open overlap does\n   not treat `[0,10)` and `[10,20)` as overlapping.\n\nRun the local validator first:\n\n```bash\npython3 -B scripts\u002Fbed_validator.py \\\n  --input data.bed.gz \\\n  --assembly GRCh38.p14 \\\n  --chrom-sizes GRCh38.p14.chrom.sizes \\\n  --require-sorted\n```\n\n## Safe local workflow\n\n1. Inventory local files, checksums, assembly, contig dictionary, coordinate\n   system, strand policy, patient\u002Freplicate groups, and intended outputs.\n2. Validate BED\u002Ffragments and estimate work. Pilot a small synthetic file.\n3. Choose Python, CLI, or Rust from the documented surface; do not translate API\n   names by guesswork.\n4. Set hard limits for input bytes\u002Frecords\u002Ffiles, threads\u002Fjobs, memory, temporary\n   disk, output size, and wall time.\n5. Run in a dedicated output directory. Refuse collisions unless overwrite was\n   explicitly approved.\n6. Revalidate output sorting, bounds, row counts, checksums, and provenance.\n\n## Current Python core\n\nImports are from submodules, not the `gtars` top level:\n\n```python\nfrom gtars.models import Region, RegionSet\n\nquery = RegionSet.from_regions(\n    [\n        Region(chr=\"chr1\", start=100, end=200, rest=None),\n        Region(chr=\"chr1\", start=300, end=400, rest=None),\n    ],\n    strands=[\"+\", \"-\"],\n)\nuniverse = RegionSet.from_vectors(\n    [\"chr1\", \"chr1\"],\n    [150, 500],\n    [350, 600],\n)\n\ncounts = query.count_overlaps(universe)       # one count per query region\nflags = query.any_overlaps(universe)          # one bool per query region\nindices = query.find_overlaps(universe)       # indices into universe\npieces = query.intersect_all(universe)        # all intersection fragments\nfraction = query.coverage(universe)           # fraction of query bp covered\n```\n\n`RegionSet.sort()` mutates and returns `None`. Set algebra includes `reduce`,\n`setdiff`, `pintersect` (pairs by index), `concat`, `union`, `jaccard`,\n`coverage`, `overlap_coefficient`, `intersect_all`, `closest`, `cluster`, and\n`gaps`. Read `references\u002Fpython-api.md` before relying on ordering or strand.\n\nConsensus is a Python binding in a different module:\n\n```python\nfrom gtars.genomic_distributions import consensus\n\nrows = consensus([query, universe])\n# rows: [{\"chr\": ..., \"start\": ..., \"end\": ..., \"count\": ...}, ...]\n```\n\nSignal-track generation is **not** exposed as `gtars.uniwig` in Python 0.9.2;\nuse the reviewed CLI or Rust API. `RegionSet.coverage()` is a base-pair set metric,\nnot a WIG\u002FbigWig generator.\n\n## Tokenizers, fragments, and reference stores\n\nUse only local constructors by default:\n\n```python\nfrom gtars.models import RegionSet\nfrom gtars.tokenizers import Tokenizer\n\ntokenizer = Tokenizer.from_bed(\"reviewed-universe.bed\")\nregions = RegionSet(\"local-query.bed\")\ntokens = tokenizer.tokenize(regions)\nencoding = tokenizer(regions)\nids = encoding[\"input_ids\"]\n```\n\n`Tokenizer.from_pretrained(name)` contacts Hugging Face and writes its cache when\nthe argument is not an existing local directory; it exposes no revision or cache\nargument. Obtain explicit approval, fetch an immutable revision through a reviewed\nmechanism, verify checksums, then pass the local snapshot directory. See\n`references\u002Ftokenizers.md`.\n\nFor refget, prefer `RefgetStore.in_memory()` or `RefgetStore.open_local(path)`.\n`open_remote(cache_path, remote_url)` contacts a remote service, creates\u002Fuses a\nlocal cache, and performs on-demand range reads. See `references\u002Frefget.md`.\n\n## Network and cache gate\n\nNo download or cache write is implicit in this skill. Before any network-capable\nupstream call:\n\n- obtain explicit user approval for the exact host, endpoint, data, and cache;\n- allowlist HTTPS hosts and reject unreviewed redirects;\n- record immutable revision\u002Fidentifier, retrieval time, expected SHA-256 and\n  domain digest, assembly accession, size quota, and provenance;\n- disclose sensitive BED coordinates, barcodes, sample labels, and reference\n  choices that could leave the approved environment;\n- validate downloaded content as untrusted before using it.\n\nImportant side effects:\n\n- `RegionSet(path)` has HTTP support; a nonexistent local string may be treated as\n  a URL. Check that the local path exists before construction.\n- `Tokenizer.from_pretrained` may download `universe.bed.gz` into the Hugging Face\n  cache.\n- `RefgetStore.on_disk` creates\u002Fwrites a store. `open_remote` loads remote metadata\n  and enables persistence by default.\n- `gtars bbcache` creates cache directories even when constructing the client.\n  Cache\u002Fdownload commands use `BBCLIENT_CACHE` (default `~\u002F.bbcache`) and\n  `BEDBASE_API` (default `https:\u002F\u002Fapi.bedbase.org`).\n\n## Sensitive metadata and leakage\n\nGenomic intervals, rare loci, barcodes, sample names, phenotypes, and assembly\nchoices can be identifying. Keep full paths and raw coordinates out of logs;\ndefault bundled reports redact paths and emit only counts\u002Fchecksums.\n\nFreeze splits by patient\u002Fdonor first, then keep all technical and biological\nreplicates in the same split. Fit consensus sets, universes, tokenizers, scaling,\nthresholds, and QC rules on training data only. Do not create a universe from all\nsamples and then split: that leaks validation\u002Ftest locus support. Record excluded\nsamples and replicate aggregation separately.\n\n## Bundled deterministic CLIs\n\nAll six helpers reject URLs, traversal, symlinks, and special files; apply byte,\nrecord, file, coordinate, and worker caps; use no network or gtars import; and\nwrite no output files. Plans contain fixed argv templates and never launch them.\n\n```bash\npython3 -B scripts\u002Fbed_validator.py --help\npython3 -B scripts\u002Fexecution_plan.py --help\npython3 -B scripts\u002Ftokenizer_manifest.py --help\npython3 -B scripts\u002Frefget_digest_plan.py --help\npython3 -B scripts\u002Fcoverage_preflight.py --help\npython3 -B scripts\u002Fartifact_inspector.py --help\n```\n\nRun synthetic tests without bytecode:\n\n```bash\nPYTHONDONTWRITEBYTECODE=1 python3 -B -m unittest discover \\\n  -s tests\u002Fgtars -p 'test_*.py' -v\n```\n\n## Migration traps removed in 1.1\n\nDo not use stale examples containing `gtars.RegionSet`,\n`RegionSet.from_bed`, `TreeTokenizer`, `gtars.igd.build_index`,\n`gtars.uniwig.coverage_from_bed`, `gtars.RefgetStore`, global\n`set_option`\u002F`set_log_level`, `parallel_apply`, or invented exception classes.\nCLI forms such as `uniwig generate`, `igd build`, `scoring score`, and\n`fragsplit cluster-split` are also stale for 0.9.0.\n\nUpstream's published docs and stubs have some drift (for example the older\n`GlobalRefgetStore` tutorial and incomplete 0.9.2 stubs). Prefer installed\nsignature smoke tests plus immutable tagged source when they conflict.\n\n## Bundled references\n\nThese are the only six bundled references; all links are local and present:\n\n- `references\u002Fpython-api.md` — exact Python 0.9.2 imports and behavior\n- `references\u002Foverlap.md` — overlap\u002Fcount\u002Fset algebra and consensus semantics\n- `references\u002Fcoverage.md` — uniwig, bigWig, coverage, sorting, and resources\n- `references\u002Ftokenizers.md` — tokenizer\u002Funiverse and fragment compatibility\n- `references\u002Frefget.md` — digests, stores, BEDbase, network\u002Fcache controls\n- `references\u002Fcli.md` — CLI 0.9.0 commands, features, and migrations\n\n## Agent operating procedure\n\n1. **Check the environment.** Confirm tool versions, the reference genome\u002Fannotation build and the input formats (FASTQ, BAM, VCF, h5ad).\n2. **Pin down the inputs.** Confirm formats, identifiers and parameters from the data or the user. Ask rather than guess any value that changes the result.\n3. **Run a small version first.** Run the pipeline on a small subset (one sample, one chromosome, a few thousand cells) first.\n4. **Execute the full task** using the instructions and references above.\n5. **Validate the result.** Check QC metrics, sample identities, genome build consistency and batch effects before interpreting results.\n6. **Report.** State what was run (versions, commands, parameters), what was checked, and what is still uncertain.\n\n| If this happens | Do this |\n|---|---|\n| Genome builds or identifiers do not match between inputs | Stop and harmonize (liftover, ID mapping) before continuing. |\n| A function, flag or endpoint in these instructions is missing in the installed version | Check the installed version's own documentation (`help()`, `--help`, official docs), adapt, and tell the user. Never invent an API. |\n| A required input, identifier or parameter is ambiguous | Ask the user, or state the assumption explicitly before running. |\n\n**Integrity rules**\n\n- Never fabricate results, parameters, identifiers, citations or statistics. If something cannot be run or verified, say so plainly.\n- Do not interpret biological significance beyond what the statistics support; report multiple-testing correction.\n- Treat version-specific details here as possibly outdated: confirm them against the official documentation for the installed version.\n- Ask before actions that cost money, consume shared GPUs or cloud quota, touch personal or patient data, or cannot be undone.\n\n## Related skills\n\n- `geniml`: Plans and audits local genomic-interval machine learning workflows with Geniml (0.8.4) and Gtars: validates BED files against chromosome si…\n- `polars-bio`: Python library polars-bio for genomic interval operations and bioinformatics file I\u002FO on Polars DataFrames, built on Arrow and DataFusion.\n- `genomic-coordinates`: Convert genomic intervals between coordinate conventions, normalise and compare variant representations, and detect assembly or contig-nami…\n",{"name":5,"license":15,"metadata":45,"description":7,"allowed-tools":50,"compatibility":51},{"tags":46,"version":47,"category":48,"maintainer":49},"genomic intervals, BED files, peak sets, genomics, bioinformatics","1.3","life-sciences","Kalaris Labs","Read Write Edit Bash Glob","Python bindings require Python 3.10+ and gtars 0.9.2. The Rust meta-crate and gtars-cli are 0.9.0 and require a Rust toolchain supporting Edition 2024; upstream declares no rust-version. Bundled audit CLIs use only Python 3.10+ standard library and are local\u002Fnetwork-free. Remote constructors, pretrained tokenizers, refget, and BEDbase caching require explicit network and storage approval.","main","2b63f856ce19c2a797929cacabd9e0fc936812fa","plugins\u002FKalarisLabs\u002Fresearch-agent-skills\u002Fskills\u002Fgtars",491,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins","https:\u002F\u002Fhol.org\u002Fplugins\u002Fbest-codex-plugins","https:\u002F\u002Fgithub.com\u002Fhashgraph-online","https:\u002F\u002Fhol.org","Open-source standards and developer tools for interoperable AI agents: discovery, identity, communication, payments, privacy and security.","2026-11-14 14:06:13",{"source":8,"skillName":5,"commands":63,"prompt":73,"zipAvailable":74},[64,67,70],{"tool":65,"command":66},"npx","npx skills add hashgraph-online\u002Fawesome-codex-plugins --skill gtars",{"tool":68,"command":69},"bunx","bunx skills add hashgraph-online\u002Fawesome-codex-plugins --skill gtars",{"tool":71,"command":72},"pnpm","pnpm dlx skills add hashgraph-online\u002Fawesome-codex-plugins --skill gtars","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002FKalarisLabs\u002Fresearch-agent-skills\u002Fskills\u002Fgtars\u002FSKILL.md",true,[76,82,86,90,94,98,102,108,112,116,120,124,128,132],{"path":77,"resourceType":78,"resourceTypeDesc":79,"sizeBytes":80,"sourceHtmlUrl":81},"references\u002Fcli.md","REFERENCE","参考文档",9887,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002FKalarisLabs\u002Fresearch-agent-skills\u002Fskills\u002Fgtars\u002Freferences\u002Fcli.md",{"path":83,"resourceType":78,"resourceTypeDesc":79,"sizeBytes":84,"sourceHtmlUrl":85},"references\u002Fcoverage.md",7721,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002FKalarisLabs\u002Fresearch-agent-skills\u002Fskills\u002Fgtars\u002Freferences\u002Fcoverage.md",{"path":87,"resourceType":78,"resourceTypeDesc":79,"sizeBytes":88,"sourceHtmlUrl":89},"references\u002Foverlap.md",7450,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002FKalarisLabs\u002Fresearch-agent-skills\u002Fskills\u002Fgtars\u002Freferences\u002Foverlap.md",{"path":91,"resourceType":78,"resourceTypeDesc":79,"sizeBytes":92,"sourceHtmlUrl":93},"references\u002Fpython-api.md",9588,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002FKalarisLabs\u002Fresearch-agent-skills\u002Fskills\u002Fgtars\u002Freferences\u002Fpython-api.md",{"path":95,"resourceType":78,"resourceTypeDesc":79,"sizeBytes":96,"sourceHtmlUrl":97},"references\u002Frefget.md",10540,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002FKalarisLabs\u002Fresearch-agent-skills\u002Fskills\u002Fgtars\u002Freferences\u002Frefget.md",{"path":99,"resourceType":78,"resourceTypeDesc":79,"sizeBytes":100,"sourceHtmlUrl":101},"references\u002Ftokenizers.md",8238,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002FKalarisLabs\u002Fresearch-agent-skills\u002Fskills\u002Fgtars\u002Freferences\u002Ftokenizers.md",{"path":103,"resourceType":104,"resourceTypeDesc":105,"sizeBytes":106,"sourceHtmlUrl":107},"scripts\u002F__init__.py","SCRIPT","脚本",57,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002FKalarisLabs\u002Fresearch-agent-skills\u002Fskills\u002Fgtars\u002Fscripts\u002F__init__.py",{"path":109,"resourceType":104,"resourceTypeDesc":105,"sizeBytes":110,"sourceHtmlUrl":111},"scripts\u002F_common.py",14970,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002FKalarisLabs\u002Fresearch-agent-skills\u002Fskills\u002Fgtars\u002Fscripts\u002F_common.py",{"path":113,"resourceType":104,"resourceTypeDesc":105,"sizeBytes":114,"sourceHtmlUrl":115},"scripts\u002Fartifact_inspector.py",10846,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002FKalarisLabs\u002Fresearch-agent-skills\u002Fskills\u002Fgtars\u002Fscripts\u002Fartifact_inspector.py",{"path":117,"resourceType":104,"resourceTypeDesc":105,"sizeBytes":118,"sourceHtmlUrl":119},"scripts\u002Fbed_validator.py",5432,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002FKalarisLabs\u002Fresearch-agent-skills\u002Fskills\u002Fgtars\u002Fscripts\u002Fbed_validator.py",{"path":121,"resourceType":104,"resourceTypeDesc":105,"sizeBytes":122,"sourceHtmlUrl":123},"scripts\u002Fcoverage_preflight.py",8273,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002FKalarisLabs\u002Fresearch-agent-skills\u002Fskills\u002Fgtars\u002Fscripts\u002Fcoverage_preflight.py",{"path":125,"resourceType":104,"resourceTypeDesc":105,"sizeBytes":126,"sourceHtmlUrl":127},"scripts\u002Fexecution_plan.py",11927,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002FKalarisLabs\u002Fresearch-agent-skills\u002Fskills\u002Fgtars\u002Fscripts\u002Fexecution_plan.py",{"path":129,"resourceType":104,"resourceTypeDesc":105,"sizeBytes":130,"sourceHtmlUrl":131},"scripts\u002Frefget_digest_plan.py",10983,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002FKalarisLabs\u002Fresearch-agent-skills\u002Fskills\u002Fgtars\u002Fscripts\u002Frefget_digest_plan.py",{"path":133,"resourceType":104,"resourceTypeDesc":105,"sizeBytes":134,"sourceHtmlUrl":135},"scripts\u002Ftokenizer_manifest.py",8558,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002FKalarisLabs\u002Fresearch-agent-skills\u002Fskills\u002Fgtars\u002Fscripts\u002Ftokenizer_manifest.py",[137,146,154,162,170,178],{"skillNo":138,"name":139,"nameZh":6,"description":140,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":141,"sourceHtmlUrl":142,"categories":143,"keywords":144,"publishedAt":145},"01M7DV332YQMGD8WTRJ9DZQXTB","worktable-review-with-annotations","Create, inspect, reply to, update, or resolve situated Worktable annotations. Use when feedback belongs to a Doc section, block, text range, or HTML Doc, or when the user asks to address existing review feedback and coordinate a requested artifact edit.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-review-with-annotations\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-review-with-annotations\u002FSKILL.md",[],[],"2026-11-15 14:19:17",{"skillNo":147,"name":148,"nameZh":6,"description":149,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":150,"sourceHtmlUrl":151,"categories":152,"keywords":153,"publishedAt":145},"01M7DV2YGHCZVBEB1AHYB6CGHE","worktable-find-and-synthesize-context","Find, assess, and synthesize context from Worktable Docs and Records. Use when the user asks what a workspace contains, wants workspace-backed research, needs current facts reconciled across artifacts, or requests a cited synthesis without changing workspace state.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-find-and-synthesize-context\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-find-and-synthesize-context\u002FSKILL.md",[],[],{"skillNo":155,"name":156,"nameZh":6,"description":157,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":158,"sourceHtmlUrl":159,"categories":160,"keywords":161,"publishedAt":145},"01M7DV2TQN1H11ME2SAF9MXN7V","worktable-create-or-update-docs","Create, edit, rename, or maintain narrative Worktable Docs. Use for plans, decisions, briefs, notes, research syntheses, and other durable prose when the user requests a new artifact or a safe update to an existing one.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-update-docs\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-update-docs\u002FSKILL.md",[],[],{"skillNo":163,"name":164,"nameZh":6,"description":165,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":166,"sourceHtmlUrl":167,"categories":168,"keywords":169,"publishedAt":145},"01M7DV2NE452NEA8TXATKSNAJH","worktable-create-or-manage-records","Organize repeated structured information as durable Worktable Records. Use for research datasets, inventories, directories, pipelines, trackers, and other collections whose items need identity, fields, filtering, links, validation, or individual updates.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-manage-records\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-manage-records\u002FSKILL.md",[],[],{"skillNo":171,"name":172,"nameZh":6,"description":173,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":174,"sourceHtmlUrl":175,"categories":176,"keywords":177,"publishedAt":145},"01M7DV2HENRBR2KSCVM064WH0Y","worktable-create-or-manage-interactive-html","Create or improve durable visual and interactive artifacts in Worktable. Use for dashboards, trackers, calculators, mini apps, slide presentations, and other requests that benefit from a polished HTML Doc.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-manage-interactive-html\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-manage-interactive-html\u002FSKILL.md",[],[],{"skillNo":179,"name":180,"nameZh":6,"description":181,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":182,"sourceHtmlUrl":183,"categories":184,"keywords":185,"publishedAt":145},"01M7DV2CADR5R7RQFFYPNAGR9Y","worktable-collaborate-threads","Read, answer, and hand off messages through durable Worktable threads with users or other authenticated participants. Use when the user asks to check addressed messages, reply to a participant, continue a thread, request another agent's help, wait briefly for a response, or watch for comments and messages sent to the agent.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-collaborate-threads\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-collaborate-threads\u002FSKILL.md",[],[]]