← Files Biological Sequence & Alignment ViewerARCHIVED FILE
starter-examples.json
17.7 KB · Sep 30, 2026 · 23:01 UTC
{
"schemaVersion": 1,
"pluginVersion": "0.1.43",
"downstreamQualification": {
"issue": "LSC-109",
"url": "https://linear.app/openai/issue/LSC-109/sequence-and-alignment-viewer-qualify-every-refreshed-starter-example",
"lastQualifiedPluginVersion": "0.1.26",
"qualificationComplete": false,
"status": "pending",
"qualificationReport": "LSC_109_QUALIFICATION.md",
"qualificationRecord": "lsc-109-qualification.json"
},
"examples": [
{
"id": "ena-drr037765-first-500",
"prompt": "Fetch ENA DRR037765 first 500 reads to active workspace; open and report live length range, GC, Q30, and subset provenance",
"area": "Sequence",
"source": {
"database": "ENA",
"stableIdentifiers": [
"DRR037765",
"DRR037765.fastq.gz@md5:81735432a6f578b332aae58cdbd95231"
],
"officialEndpoints": [
"https://www.ebi.ac.uk/ena/portal/api/filereport",
"https://ftp.sra.ebi.ac.uk/vol1/fastq/DRR037/DRR037765/DRR037765.fastq.gz"
],
"optionalSkill": "Life Science Research may retrieve or confirm ENA/SRA metadata; Codex verifies and saves only the official endpoint responses with its authorized workspace tools"
},
"bounds": {
"metadataBytes": 262144,
"compressedBytes": 16777216,
"decodedBytes": 67108864,
"sourceRecords": 967,
"selectedRecords": 500,
"selectionRule": "first 500 complete parsed FASTQ records in source order, canonical four-line FASTQ",
"liveTimeoutSeconds": 60
},
"viewer": {
"viewerCount": 1,
"sessionCount": 1,
"mode": "sequence",
"visibleState": [
"FASTQ overview",
"500 reads",
"read length, GC, and Q30 values from live viewer context"
]
},
"workflow": {
"acquisitionOwner": "codex",
"acquisitionMethod": "host-authorized-research-and-workspace-tools",
"openTool": "sequence.open_from_chat",
"operations": ["read the mounted FASTQ summary from live viewer state"],
"scientificQuestion": "Are the bounded real amplicon reads length-uniform and high quality?"
},
"expectedResults": {
"readCount": 500,
"totalBases": 235490,
"readLengthMin": 469,
"readLengthMax": 471,
"gcPercent": 28.8462355,
"q30Percent": 95.3976814,
"displayGcPercent": "28.8%",
"displayQ30Percent": "95.4%",
"artifactByteLength": 480372,
"artifactSha256": "46bd72991d9c9c2bf64751e88e52548d852d5fa021da4815ee6f6517a51b18b9"
},
"artifacts": {
"source": "codex-viewer-examples/DRR037765-first-500.fastq",
"sourceProvenance": "codex-viewer-examples/DRR037765-first-500.fastq.provenance.json",
"sourceProvenanceOwner": "codex",
"derived": null
},
"failureBehavior": [
"Codex host network, rate-limit, metadata-shape, filename, byte-count, MD5, gzip, parser, subset, quota, cancellation, or authorized-workspace failures open no viewer",
"never use a bundled or synthetic fallback",
"a changed authoritative MD5 is source drift requiring review"
],
"acceptableVariability": [
"retrieval timestamp and HTTP cache metadata",
"collision-safe source filename suffix on a non-clean workspace"
],
"tests": [
"src/public-example-acquisition.test.ts",
"src/public-example-acquisition.live.test.ts",
"e2e/public-starter-evidence.spec.ts"
]
},
{
"id": "uniprot-human-ras-sv1",
"prompt": "Fetch UniProt P01116/P01111/P01112 alignment; map conserved motifs to KRAS, compute distances/tree, publish Newick to workspace",
"area": "Alignment",
"source": {
"database": "UniProtKB",
"stableIdentifiers": ["P01116@SV1", "P01111@SV1", "P01112@SV1"],
"officialEndpoints": [
"https://rest.uniprot.org/uniprotkb/P01116.fasta",
"https://rest.uniprot.org/uniprotkb/P01111.fasta",
"https://rest.uniprot.org/uniprotkb/P01112.fasta"
],
"optionalSkill": "UniProt skill may retrieve or confirm reviewed entry metadata; Codex verifies only the fixed official FASTA responses and builds the pinned fixed-order center-star-compatible alignment with its authorized workspace tools before opening"
},
"bounds": {
"responseBytesPerRecord": 32768,
"recordCount": 3,
"residuesPerRecord": 189,
"alignmentEngine": "builtin-center-star",
"alignmentDynamicProgrammingCells": 72200,
"alignedColumns": 191,
"treeRows": 3,
"liveTimeoutSeconds": 60
},
"viewer": {
"viewerCount": 1,
"sessionCount": 1,
"mode": "alignment",
"visibleState": [
"three-row, 191-column aligned FASTA",
"P01116 KRAS as active reference",
"reference-coordinate mapping and conservation",
"distance matrix and graphical neighbor-joining tree"
]
},
"workflow": {
"acquisitionOwner": "codex",
"acquisitionMethod": "host-authorized-research-and-workspace-tools",
"openTool": "sequence.open_from_chat",
"operations": [
"sequence.control_viewer set_alignment_reference P01116",
"map KRAS reference residues 10-17, 30-38, 60-76, and 116-119",
"sequence.run_analysis distance-matrix",
"sequence.run_analysis build-tree neighbor-joining",
"Codex writes verified Newick and an accurately attributed provenance receipt to its authorized workspace; plugin publication requires independently authenticated roots"
],
"scientificQuestion": "Which GTPase-core motifs are conserved across human RAS paralogs, and where does the hypervariable tail diverge?"
},
"expectedResults": {
"artifactByteLength": 786,
"artifactSha256": "cb32dd89ca7855f7666fbdf3f2ff926f935b1dbc9e7f57573f884dda7e59c68f",
"reference": "P01116",
"motifs": [
"P-loop KRAS 10-17 GAGGVGKS",
"switch I KRAS 30-38 DEYDPTIED",
"switch II KRAS 60-76 GQEEYSAMRDQYMRTGE",
"NKXD region KRAS 116-119 NKCD"
],
"caaxTails": {
"P01116": "CIIM",
"P01111": "CVVM",
"P01112": "CVLS"
},
"pDistances": {
"P01116/P01111": 0.1315789474,
"P01116/P01112": 0.1368421053,
"P01111/P01112": 0.1578947368
},
"treeLeaves": ["P01116", "P01111", "P01112"],
"treeNewick": "('P01116':0.027632,('P01111':0.076316,'P01112':0.081579):0.027632);",
"treeInterpretation": "exploratory uncorrected p-distance guide tree; do not interpret the three-leaf topology as a publication phylogeny"
},
"artifacts": {
"source": "codex-viewer-examples/human-RAS-UniProt-SV1.aln-fasta",
"sourceProvenance": "codex-viewer-examples/human-RAS-UniProt-SV1.aln-fasta.provenance.json",
"sourceProvenanceOwner": "codex",
"derived": {
"format": "newick",
"destination": {
"kind": "workspace",
"writer": "codex",
"base": "opened-source",
"relativePath": "RAS-P01116-P01111-P01112-NJ.nwk"
},
"output": "codex-viewer-examples/RAS-P01116-P01111-P01112-NJ.nwk",
"provenance": "codex-viewer-examples/RAS-P01116-P01111-P01112-NJ.nwk.provenance.json",
"requiredProvenance": [
"sequence-viewer-guide-tree-v1 engine, neighbor-joining algorithm, uncorrected-p-distance model, row identity, and exploratory warning",
"accurately attributed Codex-authored receipt and verified live viewer analysis result",
"source workspace path and SHA-256",
"output SHA-256 and byte length"
]
}
},
"failureBehavior": [
"Codex host network, rate-limit, accession/header, sequence-version, digest, length, protein-motif, alignment, parser, cancellation, or authorized-workspace failures open no viewer",
"Codex workspace Newick publication is create-new and never overwrites an existing output or accurately attributed receipt; plugin publication fails closed without authenticated roots",
"never substitute Rfam, synthetic sequences, or an assumed workspace file"
],
"acceptableVariability": [
"retrieval timestamp and HTTP cache metadata",
"collision-safe acquired-source filename suffix on a non-clean workspace",
"equivalent Newick child ordering and display layout with the same leaf set and distances"
],
"tests": [
"src/public-example-acquisition.test.ts",
"src/public-example-acquisition.live.test.ts",
"src/msa/analysis.test.ts",
"src/msa/guide-tree.test.ts",
"src/workspace-export-publisher.test.ts",
"e2e/public-starter-evidence.spec.ts"
]
},
{
"id": "ncbi-nc-001416-1",
"prompt": "Fetch NCBI NC_001416.1 to active workspace; open it, map cI to OR1–OR3, and translate cI with code 11",
"area": "Sequence",
"source": {
"database": "NCBI Nuccore",
"stableIdentifiers": ["NC_001416.1", "NP_040628.1"],
"officialEndpoints": [
"https://eutils.ncbi.nlm.nih.gov/entrez/eutils/efetch.fcgi"
],
"optionalSkill": "NCBI Entrez skill may retrieve or confirm metadata; Codex verifies and saves only the fixed official EFetch response with its authorized workspace tools"
},
"bounds": {
"responseBytes": 2097152,
"recordCount": 1,
"residueCount": 48502,
"liveTimeoutSeconds": 60
},
"viewer": {
"viewerCount": 1,
"sessionCount": 1,
"mode": "sequence",
"visibleState": [
"NC_001416.1 metadata and 48,502-base sequence",
"cI CDS pinned on the reverse strand",
"OR3, OR2, and OR1 regulatory annotations",
"completed code-11 reverse-frame translation job"
]
},
"workflow": {
"acquisitionOwner": "codex",
"acquisitionMethod": "host-authorized-research-and-workspace-tools",
"openTool": "sequence.open_from_chat",
"operations": [
"sequence.control_viewer select_sequence_feature cI CDS",
"query exact cI and operator annotations",
"sequence.run_analysis translate start 37227 end 37940 frame -1 geneticCodeId 11"
],
"scientificQuestion": "How is the lambda cI repressor CDS oriented relative to OR1-OR3, and does code-aware translation reproduce its annotated protein?"
},
"expectedResults": {
"sequenceLength": 48502,
"ci": {
"location": "complement(37227..37940)",
"start": 37227,
"end": 37940,
"strand": "-",
"geneticCodeId": 11,
"proteinAccession": "NP_040628.1",
"aminoAcids": 237,
"codingSequenceSha256": "a51dec784e51f85a35d643a84820c89430b526cd9bf54a398b91c70045cc62e8",
"proteinSha256": "ec5d954fd10be8c19c920e78badc5d9e9cc281f6801e2c5fde3803c9f133f580"
},
"operators": {
"OR3": "37951..37967",
"OR2": "37974..37990",
"OR1": "37998..38014"
},
"artifactBaselineSha256": "3c624302adeeb3c00649f549903ab781b9e75bab16069ae655833d536407367f"
},
"artifacts": {
"source": "codex-viewer-examples/NC_001416.1.gb",
"sourceProvenance": "codex-viewer-examples/NC_001416.1.gb.provenance.json",
"sourceProvenanceOwner": "codex",
"derived": null
},
"failureBehavior": [
"Codex host network, rate-limit, accession.version, record completeness, length, cI/operator annotation, parser, size, cancellation, or authorized-workspace failures open no viewer",
"never infer or fabricate a missing feature and never use a bundled fallback"
],
"acceptableVariability": [
"retrieval timestamp, HTTP cache metadata, and GenBank annotation serialization",
"collision-safe source filename suffix on a non-clean workspace"
],
"tests": [
"src/public-example-acquisition.test.ts",
"src/public-example-acquisition.live.test.ts",
"src/sequence/feature-location.test.ts",
"src/sequence/translation.test.ts",
"e2e/public-starter-evidence.spec.ts"
]
}
],
"capabilityMatrix": [
{
"area": "Sequence",
"capability": "annotated record, exact compound feature, reference-space mapping, and genetic-code-aware translation",
"exampleIds": ["ncbi-nc-001416-1"],
"expectedState": "NC_001416.1 in Sequence mode with cI and OR1-OR3 visible; code-11 reverse-frame translation completes",
"artifactProvenance": "versioned NCBI record and accurately attributed Codex-authored SHA-256 receipt",
"failureBehavior": "identity or annotation drift fails before a viewer opens",
"tests": [
"public acquisition unit/live tests",
"sequence translation and feature-location tests",
"installed-host starter evidence"
]
},
{
"area": "Sequence",
"capability": "real-read FASTQ quality summary with deterministic bounded selection",
"exampleIds": ["ena-drr037765-first-500"],
"expectedState": "500-read FASTQ overview with live length range, GC, and Q30",
"artifactProvenance": "Codex-authored receipt recording ENA filename, byte count, MD5, subset rule, and canonical artifact SHA-256",
"failureBehavior": "metadata, checksum, parse, or subset drift fails before a viewer opens",
"tests": [
"public acquisition unit/live tests",
"installed-host starter evidence"
]
},
{
"area": "Alignment",
"capability": "multiple real protein records, bounded deterministic alignment, reference mapping, conservation, distance matrix, and graphical NJ tree",
"exampleIds": ["uniprot-human-ras-sv1"],
"expectedState": "three-row 191-column Alignment mode with P01116 reference and conserved-core/divergent-tail comparison",
"artifactProvenance": "Codex-authored receipt recording three UniProt response/sequence digests, the accurately identified host center-star-compatible algorithm and scores, and the pinned aligned artifact SHA-256",
"failureBehavior": "source, version, sequence, motif, engine, or alignment drift fails before a viewer opens",
"tests": [
"public acquisition unit/live tests",
"MSA analysis/tree tests",
"installed-host starter evidence"
]
},
{
"area": "Shared",
"capability": "Codex-host official-endpoint acquisition, accurately attributed workspace provenance, exact absolute-path opening, opaque handle, and exactly one viewer/session",
"exampleIds": [
"ena-drr037765-first-500",
"uniprot-human-ras-sv1",
"ncbi-nc-001416-1"
],
"expectedState": "authorized Codex-host download, one sequence.open_from_chat call with the exact absolute path, one opaque resource read, one card, and one active session",
"artifactProvenance": "Codex-authored workspace artifact and provenance receipt; the embedded app receives only an opaque resource, never a filesystem path",
"failureBehavior": "retrieval or source-validation failures open no viewer; plugin-managed acquisition and publication fail closed without independently authenticated roots",
"tests": [
"host-managed acquisition, rootless absolute-path opening, and authenticated-session/race tests",
"public bundle validator",
"installed-host starter evidence"
]
},
{
"area": "Output",
"capability": "Codex-authored create-new workspace Newick from a verified live analysis; plugin-managed publication only with independently authenticated roots",
"exampleIds": ["uniprot-human-ras-sv1"],
"expectedState": "valid Newick and accurately attributed Codex-authored provenance receipt saved beside the opened alignment",
"artifactProvenance": "verified live tree engine/parameters, source path/hash, output byte length/hash, and Codex receipt authorship",
"failureBehavior": "collision, containment, quota, disk, cancellation, or race never overwrites source or prior output; rootless plugin-managed publication fails closed",
"tests": [
"workspace export publisher and persistence tests",
"installed-host RAS starter evidence"
]
},
{
"area": "Not selected",
"capability": "edit-copy, annotations mutation, primer design, evidence/index/reference browser, large indexed input, and durable session restoration",
"exampleIds": [],
"expectedState": "covered by focused regression suites, not overloaded into the three concise starters",
"artifactProvenance": "existing feature-specific contracts",
"failureBehavior": "unchanged existing safety and integrity behavior",
"tests": [
"editing, annotation, primer, workspace-track, indexed-input, Save As, and session suites"
]
}
],
"evaluatedAlternatives": [
{
"candidate": "Rfam RF00360@15.1:seed",
"decision": "retain as a supported catalog acquisition but remove from the default portfolio",
"reason": "secure and compact at 9x132, but the snoZ107/R87 family is obscure and its C/D-box conservation requires more nuance than the former prompt conveyed"
},
{
"candidate": "Rfam RF00008@15.1:seed",
"decision": "not selected",
"reason": "iconic and bounded at 85x85, but the reviewed UniProt RAS trio adds protein breadth and a clearer conserved-core versus divergent-tail question"
},
{
"candidate": "Rfam RF00005 tRNA and RF00177 bacterial SSU",
"decision": "not selected",
"reason": "their 954-row and 99x1980 seed alignments respectively are too dense for a reliable concise starter tree workflow"
}
]
}
SHA-256: 751d1e88f94a89113601fdd8a9460d2ee2760671f194b9a5870c395144061201