[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7F481QAANVSP46FKFT5RYFA":3,"skill-same:01M7F481QAANVSP46FKFT5RYFA":110},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":18,"linkOnly":19,"markdownContent":20,"markdownBody":21,"frontMatter":22,"refName":28,"commitSha":29,"skillDirectory":30,"forksCount":13,"repositoryHtmlUrl":31,"repositoryHomepageUrl":6,"ownerHtmlUrl":32,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":33,"install":34,"resources":47},"01M7F481QAANVSP46FKFT5RYFA","primer-design",null,"Designs and audits PCR and RT-qPCR primers with Primer3, explicit thermodynamic conditions, reference-based off-target amplification searches, and traceable sequence coordinates. Use for designing primer pairs, checking existing primers, exon-junction or isoform-specific assays, variant masking, cloning tails, multiplex compatibility, and interpreting Primer-BLAST results. Includes bounded local in-silico PCR and BLAST screening; distinguishes computational candidates from experimentally validated assays.","beanspowt\u002Fbeancode","beanspowt","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F333943283?v=4","01M7F39F0GDWKS57BCKW8JJH7F","01M7F39F1B1TVPDJMS37BZQHWM",0,"MIT","https:\u002F\u002Fgithub.com\u002Fbeanspowt\u002Fbeancode\u002Fblob\u002Fmain\u002Fconfig\u002Fopencode\u002Fskills\u002Fprimer-design\u002FSKILL.md",[],[],"2026-11-16 10:05:34",false,"---\nname: primer-design\ndescription: Designs and audits PCR and RT-qPCR primers with Primer3, explicit thermodynamic conditions, reference-based off-target amplification searches, and traceable sequence coordinates. Use for designing primer pairs, checking existing primers, exon-junction or isoform-specific assays, variant masking, cloning tails, multiplex compatibility, and interpreting Primer-BLAST results. Includes bounded local in-silico PCR and BLAST screening; distinguishes computational candidates from experimentally validated assays.\nlicense: MIT\ncompatibility: Requires Python 3.11+ and primer3-py 2.3.1 for design and thermodynamics. Local exhaustive screening uses the standard library; BLAST screening additionally needs blastn and makeblastdb on PATH. Network access is needed only for installation, reference retrieval, or external Primer-BLAST.\nmetadata:\n  version: \"1.1\"\n  skill-author: K-Dense Inc.\n  last-reviewed: \"2026-10-01\"\n---\n\n# Primer design and specificity\n\nProduce candidate oligos in 5-prime-to-3-prime orientation, with the exact template,\nchemistry, intended products, and search scope behind each conclusion. Calculate\nsequence-dependent quantities with the supplied tools. A familiar gene name, good\nPrimer3 penalty, or a single BLAST alignment cannot establish primer specificity.\n\n## Choose the workflow\n\n| Request | Start here |\n| --- | --- |\n| New genomic PCR or RT-qPCR pair | Define the assay and reference; design; assess thermodynamics; screen products. |\n| Check an existing pair | Prepare pair TSV; assess both full oligos and annealing cores; screen with explicit intended coordinates. |\n| Exon junction, transcript isoform, allele discrimination | Read [design-workflows.md](references\u002Fdesign-workflows.md); supply sequence annotation before imposing constraints. |\n| Cloning\u002Fadaptor-tailed primers | Design annealing cores, append declared 5-prime tails, reassess full oligos, reconstruct the final product. |\n| Multiplex panel | Enable `--multiplex` in both thermodynamics and specificity tools to assess oligo interactions and cross-pair products. |\n| Degenerate, bisulfite, probe, or modified-base assay | Use the specialized workflow in [design-workflows.md](references\u002Fdesign-workflows.md); the bundled ordinary-DNA model is insufficient. |\n\nThe local tooling supports paired primers with unambiguous ACGT cores. Advanced assay\ntypes have substantive design and validation guidance, but are not silently reduced\nto ordinary PCR. This skill designs assays; expression normalization, experimental\ndiagnostic validation, and guide-RNA design are separate tasks.\n\n## Establish the assay contract\n\nObtain what changes the result; use explicit provisional assumptions for an exploratory\ndesign, and identify them in the report:\n\n- **Purpose and template:** genomic DNA, cDNA, plasmid, or another defined substrate;\n  target organism, accession **with version**, assembly\u002Ftranscript release, strand,\n  desired isoforms, and product-size range. Name wanted and unwanted templates.\n- **Sequence evidence:** local FASTA plus source\u002Fretrieval date and its SHA-256 hash.\n  A locus excerpt uses local coordinates; record its mapping to the full reference.\n  Include relevant paralogs, pseudogenes, alternate contigs, transcript isoforms,\n  vector backbone, and host sequence in the appropriate screen.\n- **Reaction conditions:** polymerase\u002Fbuffer, monovalent salt, total divalent salt,\n  total dNTP, and oligo concentrations. Primer3 uses mM for salts\u002FdNTP and nM for DNA.\n  Record the initial reaction concentration separately from Primer3's effective\n  annealing-oligo concentration parameter. Engine defaults are assumptions.\n- **Constraints:** target interval, allowed\u002Fexcluded binding regions, junctions,\n  variant exclusions and their source, fixed primers, tails, or multiplex membership.\n  Do not guess exon boundaries or silently substitute another assembly.\n- **Definition of an acceptable result:** relevant off-target references, amplicon\n  lengths, mismatch search limits, controls, and experimental validation appropriate\n  to the assay. There is no universal thermodynamic cutoff that validates all PCRs.\n\nUse [input-contract.md](references\u002Finput-contract.md) for file schemas and coordinate\nexamples. Copy [assay-report-template.md](assets\u002Fassay-report-template.md) into the\nanalysis directory to collect evidence. Reference retrieval may be manual or through\nan established sequence API; preserve accession\u002Fversion and verify the returned\nsequence. The supplied scripts use local files and do not submit sequences online.\n\n## Install and verify\n\nSet `SKILL_DIR` to this skill's actual installed directory. Work in a separate analysis\ndirectory so environments, reference databases, and results do not enter the skill.\n\n```bash\nuv venv --python 3.13 .venv-primer\nuv pip install --python .venv-primer\u002Fbin\u002Fpython -r \"$SKILL_DIR\u002Fassets\u002Frequirements.txt\"\n.venv-primer\u002Fbin\u002Fpython \"$SKILL_DIR\u002Fscripts\u002Fdesign_primers.py\" --help\n```\n\nUse the environment's `Scripts\u002Fpython.exe` on Windows. The design\u002Fthermodynamic\nexamples target primer3-py **2.3.1**, tested with Python **3.13**. For the optional\nBLAST engine, install NCBI BLAST+ from its official distribution and check:\n\n```bash\nblastn -version\nmakeblastdb -version\n```\n\nLocal integration checks also exercise BLAST+ 2.17.0; other releases require checking\ntheir output and search behavior before claiming equivalent coverage.\n\nRead [sources.md](references\u002Fsources.md) when updating dependencies or API assumptions.\nThe scripts record engine versions and effective settings in their JSON reports.\n\n## 1. Design candidates\n\nFor an initial functional demonstration, use the bundled synthetic sequence. It is\nnonbiological example input, not an experimentally validated assay:\n\n```bash\n.venv-primer\u002Fbin\u002Fpython \"$SKILL_DIR\u002Fscripts\u002Fdesign_primers.py\" \\\n  --template \"$SKILL_DIR\u002Fassets\u002Fdemo-template.fasta\" \\\n  --preset qpcr --config \"$SKILL_DIR\u002Fassets\u002Fqpcr-config.json\" \\\n  --output design.json --pairs-out pairs.tsv --expected-out expected.tsv\n```\n\nFor actual work, substitute the reviewed target FASTA and constraints. A multi-record\nFASTA requires `--record` with its exact ID. The `pcr` preset requests 100–1000 bp;\n`qpcr` requests 70–200 bp. Override these starting ranges in the configuration.\nThe example overrides its range to 90–180 bp and includes a specific target interval.\nMultiple `SEQUENCE_TARGET` intervals are alternatives: Primer3 flanks at least one.\nTo require coverage of every interval, supply one enclosing target and verify the\nreturned product; separate assays require separate design runs.\n\nThe tool:\n\n- Passes validated `sequence_args` and `global_args` to Primer3. Unsupported or\n  misspelled tags fail rather than silently changing the task.\n- Converts ambiguous template positions to `N` and forbids ambiguous primer bases.\n  It does not interpret lowercase sequence as a repeat mask; use explicit exclusions.\n- Optionally imports `--mask-bed` exclusions in **the supplied template's coordinates**.\n  Use BED intervals for variant\u002Frepeat masking only after validating the reference\n  mapping; it does not infer allele frequencies or convert arbitrary VCFs.\n- Verifies returned forward and reverse sequences against the template and checks\n  product lengths. Primer3's right-primer position is converted into a half-open\n  binding interval; the reverse primer is already in ordering orientation.\n- Optionally appends `--forward-tail` and `--reverse-tail` to cores. The design score\n  and core Tm do not include those tails; step 2 checks complete oligos.\n\nInspect `engine_explanations` when no candidates are found. Change a biologically\njustified constraint and rerun; do not silently relax all constraints or return an\ninvented sequence. Preserve previous reports when comparing parameter choices.\n\n`expected.tsv` describes intended products **on the design template**. If screening\nanother reference, map these coordinates to its exact record IDs and orientation.\nA cDNA product cannot be relabeled as a genomic interval across introns.\n\n## 2. Assess thermodynamics\n\nUse the **same chemistry as the design**. The following explicit values match the\nbundled design defaults; replace them together when the actual conditions differ:\n\n```bash\n.venv-primer\u002Fbin\u002Fpython \"$SKILL_DIR\u002Fscripts\u002Fcheck_thermodynamics.py\" \\\n  --pairs pairs.tsv --mv-conc 50 --dv-conc 1.5 --dntp-conc 0.6 \\\n  --dna-conc 50 --temp-c 37 --output thermodynamics.json\n```\n\nThe report contains core Tm, core and full-oligo hairpins\u002Fhomodimers, full-oligo\nheterodimers, self 3-prime end stability, and both directional inter-oligo 3-prime\nend-stability calculations. Delta-G and\ndelta-H are reported in kcal\u002Fmol; delta-S in cal\u002F(mol K). `--temp-c` controls the\ntemperature for delta-G, **not** a recommended PCR annealing temperature.\n\nIf design used different Tm\u002Fsalt models, also set `--tm-method` and\n`--salt-corrections-method` to match; see the mapping in the input contract.\n\nFor a panel, add `--multiplex` to examine all unordered oligo combinations, including\nforward\u002Fforward and reverse\u002Freverse between different pairs. Distinguish a panel to\nbe combined from alternative candidates that will be tested separately.\n\nRead [thermodynamics.md](references\u002Fthermodynamics.md) before interpreting these\nvalues. Full oligos longer than 60 bases are reported as unresolved; they are never\nsilently truncated. Modified bases and degenerate mixtures require a suitable model.\nThermodynamic predictions support ranking and experimental planning, not a blanket\nclaim of primer quality.\nNonfinite Tm or a Tm at\u002Fbelow absolute zero is rejected as a calculation\u002Finput\nfailure; such a result must not be ranked as an ordinary low-Tm primer.\n\n## 3. Screen amplification products\n\nRead [specificity.md](references\u002Fspecificity.md) before making a specificity claim.\nSearch each primer on both strands and pair inward-facing binding sites. A binding\nhit alone is not an amplicon; absence of a reported hit is not proof of absence.\n\n### Exhaustive bounded local search\n\n```bash\n.venv-primer\u002Fbin\u002Fpython \"$SKILL_DIR\u002Fscripts\u002Fscreen_specificity.py\" \\\n  --pairs pairs.tsv --reference \"$SKILL_DIR\u002Fassets\u002Fdemo-template.fasta\" \\\n  --expected expected.tsv --engine exhaustive \\\n  --min-product 40 --max-product 1000 --max-mismatches 2 \\\n  --three-prime-bases 5 --max-three-prime-mismatches 0 \\\n  --output specificity.json\n```\n\nThis enumerates full-length, ungapped binding sites in a **bounded local reference**\nand checks F\u002FR, R\u002FF, F\u002FF, and R\u002FR products for each pair. Reference ambiguity is\ntreated conservatively as unresolved sequence; it cannot establish a clean result.\nThe mismatch limits are a search model, not a validated polymerase discrimination\nrule. Tight 3-prime thresholds can exclude amplifiable mismatched sites; broaden\nthe search when evaluating that uncertainty.\n\nUse `--circular RECORD_ID` for a circular molecule. Wrapped products use canonical\nstart coordinates and unwrapped ends greater than the reference length; only\nproducts spanning at most one molecule are considered. Caps on comparisons, hits,\nand products prevent unbounded work; hitting a cap makes the result incomplete.\n\n### BLAST-assisted local search\n\n```bash\n.venv-primer\u002Fbin\u002Fpython \"$SKILL_DIR\u002Fscripts\u002Fscreen_specificity.py\" \\\n  --pairs pairs.tsv --reference \"$SKILL_DIR\u002Fassets\u002Fdemo-template.fasta\" \\\n  --expected expected.tsv --engine blast \\\n  --min-product 40 --max-product 1000 --max-mismatches 2 \\\n  --three-prime-bases 5 --max-three-prime-mismatches 0 \\\n  --output specificity-blast.json\n```\n\nThe script builds a temporary local BLAST database, runs short-query alignment,\nrechecks complete primer-length candidate sites, and records commands and versions.\nIt checks configured hit limits and tool failures. BLAST discovery is heuristic:\nsuccessful execution and unsaturated limits do not establish exhaustive coverage.\nThe reference is loaded in memory; plan memory and search bounds for large genomes.\nThe small example verifies execution, not human-genome sensitivity or scalability.\n\nFor broader public-reference screening, follow the official NCBI Primer-BLAST\nworkflow in [specificity.md](references\u002Fspecificity.md), choosing the organism,\ndatabase, intended templates, maximum product length, and mismatch policy deliberately.\nSubmitting a sequence sends it to NCBI; keep local-only work local. Primer-BLAST\nis not a documented REST API implemented by these scripts.\n\n### Interpret the result\n\n- `potential_off_target`: inspect every unexpected product, including its orientation,\n  length, mismatch positions, and reference identity; redesign or justify its relevance.\n- `intended_target_not_found`: resolve mapping, reference, sequence, and search problems\n  before drawing specificity conclusions.\n- `no_expected_target`: results are a product inventory, without a target-specific\n  conclusion. Supply all intended product intervals for the assay.\n- `incomplete`: reference uncertainty, resource limits, or failed computation leaves\n  the conclusion unresolved. Inspect the reported reason and repeat appropriately.\n- `no_off_target_found_within_search_scope`: state the exact reference and search\n  model. Check the report's exhaustive\u002Fheuristic flags; this is not empirical validation.\n\nFor a multiplex reaction, add `--multiplex` to the specificity command as well as\nthe thermodynamics command. The specificity tool then searches every cross-pair\noligo combination, retaining the original oligo identities and tails. It treats\ncross-pair products as unintended; intentionally shared-primer designs require\nexplicit standalone combinations with reviewed expected products. The\n`--max-panel-combinations` cap defaults to 10,000 and is checked before expansion.\nThermodynamic `--multiplex` checks structures; specificity `--multiplex` checks\nreference products. Both are necessary for this panel assessment.\n\n## 4. Validate the assay and deliver evidence\n\nSelect candidates using assay purpose, coverage, specificity evidence, and chemistry,\nnot just the lowest Primer3 penalty. Preserve multiple candidates when uncertainty\nremains. Follow [design-workflows.md](references\u002Fdesign-workflows.md) for the relevant\nexperimental checks: expected product identity\u002Fsize, negative and no-template controls,\ngenomic contamination controls for RT-qPCR, and efficiency\u002Fdynamic-range assessment\nwhen quantification is intended. A single melt peak alone does not prove identity.\n\nDeliver the completed [report template](assets\u002Fassay-report-template.md), pair TSV,\ndesign\u002Fthermodynamic\u002Fspecificity JSON, and source manifest. Include:\n\n1. Exact ordering sequences, separated annealing cores and tails, primer lengths,\n   intended product coordinates\u002Fsize, and any relevant transcript\u002Fjunction mapping.\n2. Chemistry and model settings; engine versions; reference accession\u002Frelease, source,\n   retrieval date, and file hashes; explicit exclusions and rationale.\n3. Intended and potential unintended products; uncertainty from unknown bases,\n   incomplete references, heuristic searches, caps, and unsupported assay chemistry.\n4. Experimental evidence actually obtained, outstanding validation, and the reason\n   for each recommended candidate. Label untested candidates accurately.\n\n## Validation and limits of this implementation\n\nThe repository suite at `tests\u002Fprimer-design\u002F` exercises actual Primer3 calculations,\norientation and coordinate reconstruction, masking, tails, concentration handling,\noff-target product geometry, mismatch\u002Fambiguity behavior, failure states, and local\nBLAST integration when its executables are installed. Synthetic fixtures establish\nsoftware behavior; they do not validate a biological assay or prove genome-wide recall.\n\nTools return 0 when their computation completes, 1 for a completed design with no\ncandidates or incomplete long-oligo thermodynamics, and 2 for invalid input or tool\nfailure. Read the JSON scientific status even after exit 0: finding an off-target is\na successfully completed calculation. See the input contract for screen-specific exits.\n\n## Citing Scientific Agent Skills\n\nIf used in published work, cite the upstream methods in [sources.md](references\u002Fsources.md)\nand [Scientific Agent Skills](https:\u002F\u002Farxiv.org\u002Fabs\u002F2609.00065). Report the software\nversions and assay-specific settings required to reproduce the actual analysis.\n","# Primer design and specificity\n\nProduce candidate oligos in 5-prime-to-3-prime orientation, with the exact template,\nchemistry, intended products, and search scope behind each conclusion. Calculate\nsequence-dependent quantities with the supplied tools. A familiar gene name, good\nPrimer3 penalty, or a single BLAST alignment cannot establish primer specificity.\n\n## Choose the workflow\n\n| Request | Start here |\n| --- | --- |\n| New genomic PCR or RT-qPCR pair | Define the assay and reference; design; assess thermodynamics; screen products. |\n| Check an existing pair | Prepare pair TSV; assess both full oligos and annealing cores; screen with explicit intended coordinates. |\n| Exon junction, transcript isoform, allele discrimination | Read [design-workflows.md](references\u002Fdesign-workflows.md); supply sequence annotation before imposing constraints. |\n| Cloning\u002Fadaptor-tailed primers | Design annealing cores, append declared 5-prime tails, reassess full oligos, reconstruct the final product. |\n| Multiplex panel | Enable `--multiplex` in both thermodynamics and specificity tools to assess oligo interactions and cross-pair products. |\n| Degenerate, bisulfite, probe, or modified-base assay | Use the specialized workflow in [design-workflows.md](references\u002Fdesign-workflows.md); the bundled ordinary-DNA model is insufficient. |\n\nThe local tooling supports paired primers with unambiguous ACGT cores. Advanced assay\ntypes have substantive design and validation guidance, but are not silently reduced\nto ordinary PCR. This skill designs assays; expression normalization, experimental\ndiagnostic validation, and guide-RNA design are separate tasks.\n\n## Establish the assay contract\n\nObtain what changes the result; use explicit provisional assumptions for an exploratory\ndesign, and identify them in the report:\n\n- **Purpose and template:** genomic DNA, cDNA, plasmid, or another defined substrate;\n  target organism, accession **with version**, assembly\u002Ftranscript release, strand,\n  desired isoforms, and product-size range. Name wanted and unwanted templates.\n- **Sequence evidence:** local FASTA plus source\u002Fretrieval date and its SHA-256 hash.\n  A locus excerpt uses local coordinates; record its mapping to the full reference.\n  Include relevant paralogs, pseudogenes, alternate contigs, transcript isoforms,\n  vector backbone, and host sequence in the appropriate screen.\n- **Reaction conditions:** polymerase\u002Fbuffer, monovalent salt, total divalent salt,\n  total dNTP, and oligo concentrations. Primer3 uses mM for salts\u002FdNTP and nM for DNA.\n  Record the initial reaction concentration separately from Primer3's effective\n  annealing-oligo concentration parameter. Engine defaults are assumptions.\n- **Constraints:** target interval, allowed\u002Fexcluded binding regions, junctions,\n  variant exclusions and their source, fixed primers, tails, or multiplex membership.\n  Do not guess exon boundaries or silently substitute another assembly.\n- **Definition of an acceptable result:** relevant off-target references, amplicon\n  lengths, mismatch search limits, controls, and experimental validation appropriate\n  to the assay. There is no universal thermodynamic cutoff that validates all PCRs.\n\nUse [input-contract.md](references\u002Finput-contract.md) for file schemas and coordinate\nexamples. Copy [assay-report-template.md](assets\u002Fassay-report-template.md) into the\nanalysis directory to collect evidence. Reference retrieval may be manual or through\nan established sequence API; preserve accession\u002Fversion and verify the returned\nsequence. The supplied scripts use local files and do not submit sequences online.\n\n## Install and verify\n\nSet `SKILL_DIR` to this skill's actual installed directory. Work in a separate analysis\ndirectory so environments, reference databases, and results do not enter the skill.\n\n```bash\nuv venv --python 3.13 .venv-primer\nuv pip install --python .venv-primer\u002Fbin\u002Fpython -r \"$SKILL_DIR\u002Fassets\u002Frequirements.txt\"\n.venv-primer\u002Fbin\u002Fpython \"$SKILL_DIR\u002Fscripts\u002Fdesign_primers.py\" --help\n```\n\nUse the environment's `Scripts\u002Fpython.exe` on Windows. The design\u002Fthermodynamic\nexamples target primer3-py **2.3.1**, tested with Python **3.13**. For the optional\nBLAST engine, install NCBI BLAST+ from its official distribution and check:\n\n```bash\nblastn -version\nmakeblastdb -version\n```\n\nLocal integration checks also exercise BLAST+ 2.17.0; other releases require checking\ntheir output and search behavior before claiming equivalent coverage.\n\nRead [sources.md](references\u002Fsources.md) when updating dependencies or API assumptions.\nThe scripts record engine versions and effective settings in their JSON reports.\n\n## 1. Design candidates\n\nFor an initial functional demonstration, use the bundled synthetic sequence. It is\nnonbiological example input, not an experimentally validated assay:\n\n```bash\n.venv-primer\u002Fbin\u002Fpython \"$SKILL_DIR\u002Fscripts\u002Fdesign_primers.py\" \\\n  --template \"$SKILL_DIR\u002Fassets\u002Fdemo-template.fasta\" \\\n  --preset qpcr --config \"$SKILL_DIR\u002Fassets\u002Fqpcr-config.json\" \\\n  --output design.json --pairs-out pairs.tsv --expected-out expected.tsv\n```\n\nFor actual work, substitute the reviewed target FASTA and constraints. A multi-record\nFASTA requires `--record` with its exact ID. The `pcr` preset requests 100–1000 bp;\n`qpcr` requests 70–200 bp. Override these starting ranges in the configuration.\nThe example overrides its range to 90–180 bp and includes a specific target interval.\nMultiple `SEQUENCE_TARGET` intervals are alternatives: Primer3 flanks at least one.\nTo require coverage of every interval, supply one enclosing target and verify the\nreturned product; separate assays require separate design runs.\n\nThe tool:\n\n- Passes validated `sequence_args` and `global_args` to Primer3. Unsupported or\n  misspelled tags fail rather than silently changing the task.\n- Converts ambiguous template positions to `N` and forbids ambiguous primer bases.\n  It does not interpret lowercase sequence as a repeat mask; use explicit exclusions.\n- Optionally imports `--mask-bed` exclusions in **the supplied template's coordinates**.\n  Use BED intervals for variant\u002Frepeat masking only after validating the reference\n  mapping; it does not infer allele frequencies or convert arbitrary VCFs.\n- Verifies returned forward and reverse sequences against the template and checks\n  product lengths. Primer3's right-primer position is converted into a half-open\n  binding interval; the reverse primer is already in ordering orientation.\n- Optionally appends `--forward-tail` and `--reverse-tail` to cores. The design score\n  and core Tm do not include those tails; step 2 checks complete oligos.\n\nInspect `engine_explanations` when no candidates are found. Change a biologically\njustified constraint and rerun; do not silently relax all constraints or return an\ninvented sequence. Preserve previous reports when comparing parameter choices.\n\n`expected.tsv` describes intended products **on the design template**. If screening\nanother reference, map these coordinates to its exact record IDs and orientation.\nA cDNA product cannot be relabeled as a genomic interval across introns.\n\n## 2. Assess thermodynamics\n\nUse the **same chemistry as the design**. The following explicit values match the\nbundled design defaults; replace them together when the actual conditions differ:\n\n```bash\n.venv-primer\u002Fbin\u002Fpython \"$SKILL_DIR\u002Fscripts\u002Fcheck_thermodynamics.py\" \\\n  --pairs pairs.tsv --mv-conc 50 --dv-conc 1.5 --dntp-conc 0.6 \\\n  --dna-conc 50 --temp-c 37 --output thermodynamics.json\n```\n\nThe report contains core Tm, core and full-oligo hairpins\u002Fhomodimers, full-oligo\nheterodimers, self 3-prime end stability, and both directional inter-oligo 3-prime\nend-stability calculations. Delta-G and\ndelta-H are reported in kcal\u002Fmol; delta-S in cal\u002F(mol K). `--temp-c` controls the\ntemperature for delta-G, **not** a recommended PCR annealing temperature.\n\nIf design used different Tm\u002Fsalt models, also set `--tm-method` and\n`--salt-corrections-method` to match; see the mapping in the input contract.\n\nFor a panel, add `--multiplex` to examine all unordered oligo combinations, including\nforward\u002Fforward and reverse\u002Freverse between different pairs. Distinguish a panel to\nbe combined from alternative candidates that will be tested separately.\n\nRead [thermodynamics.md](references\u002Fthermodynamics.md) before interpreting these\nvalues. Full oligos longer than 60 bases are reported as unresolved; they are never\nsilently truncated. Modified bases and degenerate mixtures require a suitable model.\nThermodynamic predictions support ranking and experimental planning, not a blanket\nclaim of primer quality.\nNonfinite Tm or a Tm at\u002Fbelow absolute zero is rejected as a calculation\u002Finput\nfailure; such a result must not be ranked as an ordinary low-Tm primer.\n\n## 3. Screen amplification products\n\nRead [specificity.md](references\u002Fspecificity.md) before making a specificity claim.\nSearch each primer on both strands and pair inward-facing binding sites. A binding\nhit alone is not an amplicon; absence of a reported hit is not proof of absence.\n\n### Exhaustive bounded local search\n\n```bash\n.venv-primer\u002Fbin\u002Fpython \"$SKILL_DIR\u002Fscripts\u002Fscreen_specificity.py\" \\\n  --pairs pairs.tsv --reference \"$SKILL_DIR\u002Fassets\u002Fdemo-template.fasta\" \\\n  --expected expected.tsv --engine exhaustive \\\n  --min-product 40 --max-product 1000 --max-mismatches 2 \\\n  --three-prime-bases 5 --max-three-prime-mismatches 0 \\\n  --output specificity.json\n```\n\nThis enumerates full-length, ungapped binding sites in a **bounded local reference**\nand checks F\u002FR, R\u002FF, F\u002FF, and R\u002FR products for each pair. Reference ambiguity is\ntreated conservatively as unresolved sequence; it cannot establish a clean result.\nThe mismatch limits are a search model, not a validated polymerase discrimination\nrule. Tight 3-prime thresholds can exclude amplifiable mismatched sites; broaden\nthe search when evaluating that uncertainty.\n\nUse `--circular RECORD_ID` for a circular molecule. Wrapped products use canonical\nstart coordinates and unwrapped ends greater than the reference length; only\nproducts spanning at most one molecule are considered. Caps on comparisons, hits,\nand products prevent unbounded work; hitting a cap makes the result incomplete.\n\n### BLAST-assisted local search\n\n```bash\n.venv-primer\u002Fbin\u002Fpython \"$SKILL_DIR\u002Fscripts\u002Fscreen_specificity.py\" \\\n  --pairs pairs.tsv --reference \"$SKILL_DIR\u002Fassets\u002Fdemo-template.fasta\" \\\n  --expected expected.tsv --engine blast \\\n  --min-product 40 --max-product 1000 --max-mismatches 2 \\\n  --three-prime-bases 5 --max-three-prime-mismatches 0 \\\n  --output specificity-blast.json\n```\n\nThe script builds a temporary local BLAST database, runs short-query alignment,\nrechecks complete primer-length candidate sites, and records commands and versions.\nIt checks configured hit limits and tool failures. BLAST discovery is heuristic:\nsuccessful execution and unsaturated limits do not establish exhaustive coverage.\nThe reference is loaded in memory; plan memory and search bounds for large genomes.\nThe small example verifies execution, not human-genome sensitivity or scalability.\n\nFor broader public-reference screening, follow the official NCBI Primer-BLAST\nworkflow in [specificity.md](references\u002Fspecificity.md), choosing the organism,\ndatabase, intended templates, maximum product length, and mismatch policy deliberately.\nSubmitting a sequence sends it to NCBI; keep local-only work local. Primer-BLAST\nis not a documented REST API implemented by these scripts.\n\n### Interpret the result\n\n- `potential_off_target`: inspect every unexpected product, including its orientation,\n  length, mismatch positions, and reference identity; redesign or justify its relevance.\n- `intended_target_not_found`: resolve mapping, reference, sequence, and search problems\n  before drawing specificity conclusions.\n- `no_expected_target`: results are a product inventory, without a target-specific\n  conclusion. Supply all intended product intervals for the assay.\n- `incomplete`: reference uncertainty, resource limits, or failed computation leaves\n  the conclusion unresolved. Inspect the reported reason and repeat appropriately.\n- `no_off_target_found_within_search_scope`: state the exact reference and search\n  model. Check the report's exhaustive\u002Fheuristic flags; this is not empirical validation.\n\nFor a multiplex reaction, add `--multiplex` to the specificity command as well as\nthe thermodynamics command. The specificity tool then searches every cross-pair\noligo combination, retaining the original oligo identities and tails. It treats\ncross-pair products as unintended; intentionally shared-primer designs require\nexplicit standalone combinations with reviewed expected products. The\n`--max-panel-combinations` cap defaults to 10,000 and is checked before expansion.\nThermodynamic `--multiplex` checks structures; specificity `--multiplex` checks\nreference products. Both are necessary for this panel assessment.\n\n## 4. Validate the assay and deliver evidence\n\nSelect candidates using assay purpose, coverage, specificity evidence, and chemistry,\nnot just the lowest Primer3 penalty. Preserve multiple candidates when uncertainty\nremains. Follow [design-workflows.md](references\u002Fdesign-workflows.md) for the relevant\nexperimental checks: expected product identity\u002Fsize, negative and no-template controls,\ngenomic contamination controls for RT-qPCR, and efficiency\u002Fdynamic-range assessment\nwhen quantification is intended. A single melt peak alone does not prove identity.\n\nDeliver the completed [report template](assets\u002Fassay-report-template.md), pair TSV,\ndesign\u002Fthermodynamic\u002Fspecificity JSON, and source manifest. Include:\n\n1. Exact ordering sequences, separated annealing cores and tails, primer lengths,\n   intended product coordinates\u002Fsize, and any relevant transcript\u002Fjunction mapping.\n2. Chemistry and model settings; engine versions; reference accession\u002Frelease, source,\n   retrieval date, and file hashes; explicit exclusions and rationale.\n3. Intended and potential unintended products; uncertainty from unknown bases,\n   incomplete references, heuristic searches, caps, and unsupported assay chemistry.\n4. Experimental evidence actually obtained, outstanding validation, and the reason\n   for each recommended candidate. Label untested candidates accurately.\n\n## Validation and limits of this implementation\n\nThe repository suite at `tests\u002Fprimer-design\u002F` exercises actual Primer3 calculations,\norientation and coordinate reconstruction, masking, tails, concentration handling,\noff-target product geometry, mismatch\u002Fambiguity behavior, failure states, and local\nBLAST integration when its executables are installed. Synthetic fixtures establish\nsoftware behavior; they do not validate a biological assay or prove genome-wide recall.\n\nTools return 0 when their computation completes, 1 for a completed design with no\ncandidates or incomplete long-oligo thermodynamics, and 2 for invalid input or tool\nfailure. Read the JSON scientific status even after exit 0: finding an off-target is\na successfully completed calculation. See the input contract for screen-specific exits.\n\n## Citing Scientific Agent Skills\n\nIf used in published work, cite the upstream methods in [sources.md](references\u002Fsources.md)\nand [Scientific Agent Skills](https:\u002F\u002Farxiv.org\u002Fabs\u002F2609.00065). Report the software\nversions and assay-specific settings required to reproduce the actual analysis.\n",{"name":5,"license":14,"metadata":23,"description":7,"compatibility":27},{"version":24,"skill-author":25,"last-reviewed":26},"1.1","K-Dense Inc.","2026-10-01","Requires Python 3.11+ and primer3-py 2.3.1 for design and thermodynamics. Local exhaustive screening uses the standard library; BLAST screening additionally needs blastn and makeblastdb on PATH. Network access is needed only for installation, reference retrieval, or external Primer-BLAST.","main","66c9bd32ac14a921980362cf85872a1c402101c2","config\u002Fopencode\u002Fskills\u002Fprimer-design","https:\u002F\u002Fgithub.com\u002Fbeanspowt\u002Fbeancode","https:\u002F\u002Fgithub.com\u002Fbeanspowt","2026-11-10 23:22:33",{"source":8,"skillName":5,"commands":35,"prompt":45,"zipAvailable":46},[36,39,42],{"tool":37,"command":38},"npx","npx skills add beanspowt\u002Fbeancode --skill primer-design",{"tool":40,"command":41},"bunx","bunx skills add beanspowt\u002Fbeancode --skill primer-design",{"tool":43,"command":44},"pnpm","pnpm dlx skills add beanspowt\u002Fbeancode --skill primer-design","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fbeanspowt\u002Fbeancode\u002Fblob\u002Fmain\u002Fconfig\u002Fopencode\u002Fskills\u002Fprimer-design\u002FSKILL.md",true,[48,54,58,62,66,72,76,80,84,88,94,98,102,106],{"path":49,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":52,"sourceHtmlUrl":53},"assets\u002Fassay-report-template.md","ASSET","资产文件",2440,"https:\u002F\u002Fgithub.com\u002Fbeanspowt\u002Fbeancode\u002Fblob\u002Fmain\u002Fconfig\u002Fopencode\u002Fskills\u002Fprimer-design\u002Fassets\u002Fassay-report-template.md",{"path":55,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":56,"sourceHtmlUrl":57},"assets\u002Fdemo-template.fasta",577,"https:\u002F\u002Fgithub.com\u002Fbeanspowt\u002Fbeancode\u002Fblob\u002Fmain\u002Fconfig\u002Fopencode\u002Fskills\u002Fprimer-design\u002Fassets\u002Fdemo-template.fasta",{"path":59,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":60,"sourceHtmlUrl":61},"assets\u002Fqpcr-config.json",233,"https:\u002F\u002Fgithub.com\u002Fbeanspowt\u002Fbeancode\u002Fblob\u002Fmain\u002Fconfig\u002Fopencode\u002Fskills\u002Fprimer-design\u002Fassets\u002Fqpcr-config.json",{"path":63,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":64,"sourceHtmlUrl":65},"assets\u002Frequirements.txt",18,"https:\u002F\u002Fgithub.com\u002Fbeanspowt\u002Fbeancode\u002Fblob\u002Fmain\u002Fconfig\u002Fopencode\u002Fskills\u002Fprimer-design\u002Fassets\u002Frequirements.txt",{"path":67,"resourceType":68,"resourceTypeDesc":69,"sizeBytes":70,"sourceHtmlUrl":71},"references\u002Fdesign-workflows.md","REFERENCE","参考文档",10570,"https:\u002F\u002Fgithub.com\u002Fbeanspowt\u002Fbeancode\u002Fblob\u002Fmain\u002Fconfig\u002Fopencode\u002Fskills\u002Fprimer-design\u002Freferences\u002Fdesign-workflows.md",{"path":73,"resourceType":68,"resourceTypeDesc":69,"sizeBytes":74,"sourceHtmlUrl":75},"references\u002Finput-contract.md",11910,"https:\u002F\u002Fgithub.com\u002Fbeanspowt\u002Fbeancode\u002Fblob\u002Fmain\u002Fconfig\u002Fopencode\u002Fskills\u002Fprimer-design\u002Freferences\u002Finput-contract.md",{"path":77,"resourceType":68,"resourceTypeDesc":69,"sizeBytes":78,"sourceHtmlUrl":79},"references\u002Fsources.md",4552,"https:\u002F\u002Fgithub.com\u002Fbeanspowt\u002Fbeancode\u002Fblob\u002Fmain\u002Fconfig\u002Fopencode\u002Fskills\u002Fprimer-design\u002Freferences\u002Fsources.md",{"path":81,"resourceType":68,"resourceTypeDesc":69,"sizeBytes":82,"sourceHtmlUrl":83},"references\u002Fspecificity.md",10559,"https:\u002F\u002Fgithub.com\u002Fbeanspowt\u002Fbeancode\u002Fblob\u002Fmain\u002Fconfig\u002Fopencode\u002Fskills\u002Fprimer-design\u002Freferences\u002Fspecificity.md",{"path":85,"resourceType":68,"resourceTypeDesc":69,"sizeBytes":86,"sourceHtmlUrl":87},"references\u002Fthermodynamics.md",7172,"https:\u002F\u002Fgithub.com\u002Fbeanspowt\u002Fbeancode\u002Fblob\u002Fmain\u002Fconfig\u002Fopencode\u002Fskills\u002Fprimer-design\u002Freferences\u002Fthermodynamics.md",{"path":89,"resourceType":90,"resourceTypeDesc":91,"sizeBytes":92,"sourceHtmlUrl":93},"scripts\u002F_common.py","SCRIPT","脚本",5626,"https:\u002F\u002Fgithub.com\u002Fbeanspowt\u002Fbeancode\u002Fblob\u002Fmain\u002Fconfig\u002Fopencode\u002Fskills\u002Fprimer-design\u002Fscripts\u002F_common.py",{"path":95,"resourceType":90,"resourceTypeDesc":91,"sizeBytes":96,"sourceHtmlUrl":97},"scripts\u002F_specificity.py",20595,"https:\u002F\u002Fgithub.com\u002Fbeanspowt\u002Fbeancode\u002Fblob\u002Fmain\u002Fconfig\u002Fopencode\u002Fskills\u002Fprimer-design\u002Fscripts\u002F_specificity.py",{"path":99,"resourceType":90,"resourceTypeDesc":91,"sizeBytes":100,"sourceHtmlUrl":101},"scripts\u002Fcheck_thermodynamics.py",8742,"https:\u002F\u002Fgithub.com\u002Fbeanspowt\u002Fbeancode\u002Fblob\u002Fmain\u002Fconfig\u002Fopencode\u002Fskills\u002Fprimer-design\u002Fscripts\u002Fcheck_thermodynamics.py",{"path":103,"resourceType":90,"resourceTypeDesc":91,"sizeBytes":104,"sourceHtmlUrl":105},"scripts\u002Fdesign_primers.py",15676,"https:\u002F\u002Fgithub.com\u002Fbeanspowt\u002Fbeancode\u002Fblob\u002Fmain\u002Fconfig\u002Fopencode\u002Fskills\u002Fprimer-design\u002Fscripts\u002Fdesign_primers.py",{"path":107,"resourceType":90,"resourceTypeDesc":91,"sizeBytes":108,"sourceHtmlUrl":109},"scripts\u002Fscreen_specificity.py",11007,"https:\u002F\u002Fgithub.com\u002Fbeanspowt\u002Fbeancode\u002Fblob\u002Fmain\u002Fconfig\u002Fopencode\u002Fskills\u002Fprimer-design\u002Fscripts\u002Fscreen_specificity.py",[111,118,125,132,139,146],{"skillNo":112,"name":113,"nameZh":6,"description":114,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":115,"categories":116,"keywords":117,"publishedAt":18},"01M7F6BB3VS43C65P23JR4ZN5G","venue-templates","Prepares journal manuscripts, conference papers, research posters, and grant documents using venue-specific formatting guidance and bundled LaTeX scaffolds. Supports selecting an official template, checking current page or anonymity rules, adapting academic writing to a venue, or inspecting a submission PDF.","https:\u002F\u002Fgithub.com\u002Fbeanspowt\u002Fbeancode\u002Fblob\u002Fmain\u002Fhermes\u002Fprofiles\u002Fworker-writer\u002Fskills\u002Fvenue-templates\u002FSKILL.md",[],[],{"skillNo":119,"name":120,"nameZh":6,"description":121,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":122,"categories":123,"keywords":124,"publishedAt":18},"01M7F6B4CBVEHY1RKVW55X8484","scientific-writing","Drafts, revises, and audits scientific manuscripts or reports with explicit evidence provenance, reporting-guideline coverage, authorship accountability, confidentiality controls, and local consistency checks. Use for manuscript sections, references, declarations, tables, figures, or submission preparation when scientific accuracy and traceability matter.","https:\u002F\u002Fgithub.com\u002Fbeanspowt\u002Fbeancode\u002Fblob\u002Fmain\u002Fhermes\u002Fprofiles\u002Fworker-writer\u002Fskills\u002Fscientific-writing\u002FSKILL.md",[],[],{"skillNo":126,"name":127,"nameZh":6,"description":128,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":129,"categories":130,"keywords":131,"publishedAt":18},"01M7F6AXKSP5AXE8B7RCE1KP0C","scientific-visualization","Creates and audits truthful, accessible, publication-ready scientific figures with Matplotlib, Seaborn, or Plotly. Use it for figure design, multi-panel layouts, uncertainty and missing-data displays, color\u002Fcontrast review, image metadata validation, and journal export planning.","https:\u002F\u002Fgithub.com\u002Fbeanspowt\u002Fbeancode\u002Fblob\u002Fmain\u002Fhermes\u002Fprofiles\u002Fworker-writer\u002Fskills\u002Fscientific-visualization\u002FSKILL.md",[],[],{"skillNo":133,"name":134,"nameZh":6,"description":135,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":136,"categories":137,"keywords":138,"publishedAt":18},"01M7F6AMRHS0QH64CMKE8TP9Y7","scientific-slides","Builds slide decks and presentations for research talks. Used for making PowerPoint slides, conference presentations, seminar talks, research presentations, thesis defense slides, or any scientific talk. Provides slide structure, design templates, timing guidance, and visual validation. Works with PowerPoint and LaTeX Beamer.","https:\u002F\u002Fgithub.com\u002Fbeanspowt\u002Fbeancode\u002Fblob\u002Fmain\u002Fhermes\u002Fprofiles\u002Fworker-writer\u002Fskills\u002Fscientific-slides\u002FSKILL.md",[],[],{"skillNo":140,"name":141,"nameZh":6,"description":142,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":143,"categories":144,"keywords":145,"publishedAt":18},"01M7F6ACZ7GQJNNW0682235MC3","scientific-schematics","Generates scientific diagram drafts using Nano Banana 2 AI with smart iterative refinement. Uses Gemini 3.7 Flash for quality review. Refines when the review requests improvement, with at most two generations. Specialized in neural network architectures, system diagrams, flowcharts, biological pathways, and complex scientific visualizations.","https:\u002F\u002Fgithub.com\u002Fbeanspowt\u002Fbeancode\u002Fblob\u002Fmain\u002Fhermes\u002Fprofiles\u002Fworker-writer\u002Fskills\u002Fscientific-schematics\u002FSKILL.md",[],[],{"skillNo":147,"name":148,"nameZh":6,"description":149,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":150,"categories":151,"keywords":152,"publishedAt":18},"01M7F6A2JDHTYABYSA4NCE82VG","research-grants","Supports research proposal preparation and review for NSF, NIH, DOE, DARPA, and Taiwan NSTC, including opportunity-specific requirements, aims, review criteria, budgets, broader impacts, forms, and resubmissions. Use for investigator-authored grant development, compliance matrices, and proposal critiques.","https:\u002F\u002Fgithub.com\u002Fbeanspowt\u002Fbeancode\u002Fblob\u002Fmain\u002Fhermes\u002Fprofiles\u002Fworker-writer\u002Fskills\u002Fresearch-grants\u002FSKILL.md",[],[]]