encode-toolkit 0.3.0b6__tar.gz → 0.3.0b9__tar.gz

encode_toolkit-0.3.0b9/.claude-plugin/marketplace.json

@@ -0,0 +1,18 @@
+{
+  "name": "ammawla",
+  "owner": {
+    "name": "Dr. Alex M. Mawla, PhD"
+  },
+  "metadata": {
+    "description": "ENCODE Project genomics research infrastructure for Claude Code",
+    "homepage": "https://github.com/ammawla/encode-toolkit"
+  },
+  "plugins": [
+    {
+      "name": "encode-toolkit",
+      "version": "0.3.0-beta.9",
+      "source": "./plugin",
+      "description": "20 ENCODE API tools + 47 expert skills for genomics research. Search experiments, download files with MD5 verification, run pipelines, and cross-reference 14 databases."
+    }
+  ]
+}

{encode_toolkit-0.3.0b6 → encode_toolkit-0.3.0b9}/.claude-plugin/plugin.json

@@ -1,7 +1,7 @@
 {
   "name": "encode-toolkit",
-  "description": "20 ENCODE API tools + 48 expert skills for genomics research. Search experiments, download files with MD5 verification, run pipelines, and cross-reference 14 databases.",
-  "version": "0.3.0-beta.6",
+  "description": "20 ENCODE API tools + 47 expert skills for genomics research. Search experiments, download files with MD5 verification, run pipelines, and cross-reference 14 databases.",
+  "version": "0.3.0-beta.9",
   "author": {
     "name": "Dr. Alex M. Mawla, PhD",
     "email": "ammawla@ucdavis.edu"
@@ -35,7 +35,12 @@
     "biology",
     "science"
   ],
-  "skills": "skills/*",
+  "mcpServers": {
+    "encode-toolkit": {
+      "command": "npx",
+      "args": ["-y", "encode-toolkit@latest"]
+    }
+  },
   "tools": [
     {
       "name": "encode_search_experiments",

encode_toolkit-0.3.0b9/.cursor-plugin/marketplace.json

@@ -0,0 +1,27 @@
+{
+  "name": "ammawla",
+  "owner": {
+    "name": "Dr. Alex M. Mawla, PhD",
+    "email": "ammawla@ucdavis.edu"
+  },
+  "metadata": {
+    "description": "ENCODE Project genomics research infrastructure for Cursor",
+    "version": "0.3.0-beta.9",
+    "homepage": "https://github.com/ammawla/encode-toolkit"
+  },
+  "plugins": [
+    {
+      "name": "encode-toolkit",
+      "source": ".",
+      "description": "20 ENCODE API tools + 47 expert skills for genomics research. Search experiments, download files with MD5 verification, run pipelines, and cross-reference 14 databases.",
+      "category": "science",
+      "tags": [
+        "genomics",
+        "bioinformatics",
+        "encode",
+        "epigenomics",
+        "pipelines"
+      ]
+    }
+  ]
+}

encode_toolkit-0.3.0b9/.cursor-plugin/plugin.json

@@ -0,0 +1,48 @@
+{
+  "name": "encode-toolkit",
+  "description": "20 ENCODE API tools + 47 expert skills for genomics research. Search experiments, download files with MD5 verification, run pipelines, and cross-reference 14 databases.",
+  "version": "0.3.0-beta.9",
+  "author": {
+    "name": "Dr. Alex M. Mawla, PhD",
+    "email": "ammawla@ucdavis.edu"
+  },
+  "homepage": "https://github.com/ammawla/encode-toolkit",
+  "repository": "https://github.com/ammawla/encode-toolkit",
+  "license": "CC-BY-NC-ND-4.0",
+  "logo": "docs/icon.svg",
+  "keywords": [
+    "genomics",
+    "encode",
+    "bioinformatics",
+    "epigenomics",
+    "chip-seq",
+    "atac-seq",
+    "rna-seq",
+    "wgbs",
+    "hi-c",
+    "cut-and-run",
+    "pipeline",
+    "nextflow",
+    "gtex",
+    "clinvar",
+    "gwas",
+    "jaspar",
+    "cellxgene",
+    "gnomad",
+    "ensembl",
+    "ucsc",
+    "provenance",
+    "biology",
+    "science"
+  ],
+  "rules": "rules",
+  "skills": "plugin/skills",
+  "commands": "commands",
+  "agents": "agents",
+  "mcpServers": {
+    "encode-toolkit": {
+      "command": "npx",
+      "args": ["-y", "encode-toolkit@latest"]
+    }
+  }
+}

{encode_toolkit-0.3.0b6 → encode_toolkit-0.3.0b9}/.gitignore

@@ -41,28 +41,18 @@ node_modules/
 .claude/settings.local.json
 .claude/*.local.md
 
-# Data files (generated at runtime)
+# Runtime data
 *.db
 *.sqlite
 *.sqlite3
 credentials*
 data/
 
-# MCP Registry tokens
-.mcpregistry_*
+# OS
+.DS_Store
+Thumbs.db
 
-# Generated PDFs
+# Build artifacts
 *.pdf
-
-# PDF build artifacts
 docs/pdf-build/
-
-# Smithery
-.smithery/
-
-# npm lock (thin wrapper, no deps to lock)
 package-lock.json
-
-# OS
-.DS_Store
-Thumbs.db

encode_toolkit-0.3.0b9/.mcp.json

@@ -0,0 +1,3 @@
+{
+  "mcpServers": {}
+}

{encode_toolkit-0.3.0b6 → encode_toolkit-0.3.0b9}/.pre-commit-config.yaml

@@ -15,9 +15,27 @@ repos:
       - id: check-json
       - id: check-added-large-files
         args: ['--maxkb=500']
+      - id: check-case-conflict
+      - id: check-merge-conflict
       - id: no-commit-to-branch
         args: ['--branch', 'main']
 
+  - repo: local
+    hooks:
+      - id: forbid-os-files
+        name: Block OS-generated files
+        entry: >-
+          bash -c 'FOUND=$(git diff --cached --name-only --diff-filter=ACR |
+          grep -E "(\.DS_Store|Thumbs\.db|desktop\.ini)$" || true);
+          if [ -n "$FOUND" ]; then
+          echo "Blocked OS-generated files from commit:";
+          echo "$FOUND";
+          echo "Run: git rm --cached <file>";
+          exit 1; fi'
+        language: system
+        always_run: true
+        pass_filenames: false
+
   - repo: https://github.com/pre-commit/mirrors-mypy
     rev: v1.14.1
     hooks:

{encode_toolkit-0.3.0b6 → encode_toolkit-0.3.0b9}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: encode-toolkit
-Version: 0.3.0b6
+Version: 0.3.0b9
 Summary: MCP server for querying and downloading ENCODE Project genomics data directly from Claude
 Project-URL: Homepage, https://github.com/ammawla/encode-toolkit
 Project-URL: Repository, https://github.com/ammawla/encode-toolkit
@@ -37,7 +37,7 @@ Description-Content-Type: text/markdown
 
 [![License: CC BY-NC-ND 4.0](https://img.shields.io/badge/License-CC_BY--NC--ND_4.0-red.svg)](LICENSE)
 [![Python 3.10+](https://img.shields.io/badge/python-3.10+-blue.svg)](https://www.python.org/downloads/)
-[![Version](https://img.shields.io/badge/version-0.3.0--beta-yellow)](CHANGELOG.md)
+[![Version](https://img.shields.io/badge/version-0.3.0--beta.8-yellow)](CHANGELOG.md)
 [![Status](https://img.shields.io/badge/status-beta-yellow)]()
 [![Skills](https://img.shields.io/badge/skills-47-orange)](docs/skill-vignettes/)
 [![Tools](https://img.shields.io/badge/MCP_tools-20-purple)](src/encode_connector/server/main.py)
@@ -65,16 +65,31 @@ Search ENCODE, cross-reference 14 databases, run 7 analysis pipelines, and gener
 
 ## Quick Start
 
-### Claude Code (recommended)
+### Claude Code Plugin (recommended)
+
+Start a new Claude Code session and enter:
+
+```
+/plugin marketplace add ammawla/encode-toolkit
+
+/plugin install encode-toolkit
+```
+
+That's it. All 20 tools, 47 skills, and the MCP connector are now available.
+
+<details>
+<summary><strong>MCP-only install (tools only, no skills)</strong></summary>
+
+If you only need the 20 MCP tools without the 47 workflow skills:
 
 ```bash
 claude mcp add encode -- uvx encode-toolkit
 ```
 
-That's it. All 20 tools and 47 skills are now available in Claude Code.
+</details>
 
 <details>
-<summary><strong>Other installation methods</strong></summary>
+<summary><strong>Other editors and platforms</strong></summary>
 
 #### npx (Node.js)
 
@@ -102,24 +117,10 @@ Then use `encode-toolkit` as the command in any MCP client configuration:
 }
 ```
 
-#### Claude Code — Plugin Install
-
-For the full experience (20 tools + 47 skills), install as a Claude Code plugin:
-
-```bash
-claude plugin add /path/to/encode-toolkit
-```
-
-Or install from a marketplace:
-
-```
-/plugin install encode-toolkit
-```
-
 </details>
 
 <details>
-<summary><strong>Claude Desktop</strong></summary>
+<summary><strong>Claude Desktop (MCP only)</strong></summary>
 
 Add to your `claude_desktop_config.json`:
 

{encode_toolkit-0.3.0b6 → encode_toolkit-0.3.0b9}/README.md

@@ -4,7 +4,7 @@
 
 [![License: CC BY-NC-ND 4.0](https://img.shields.io/badge/License-CC_BY--NC--ND_4.0-red.svg)](LICENSE)
 [![Python 3.10+](https://img.shields.io/badge/python-3.10+-blue.svg)](https://www.python.org/downloads/)
-[![Version](https://img.shields.io/badge/version-0.3.0--beta-yellow)](CHANGELOG.md)
+[![Version](https://img.shields.io/badge/version-0.3.0--beta.8-yellow)](CHANGELOG.md)
 [![Status](https://img.shields.io/badge/status-beta-yellow)]()
 [![Skills](https://img.shields.io/badge/skills-47-orange)](docs/skill-vignettes/)
 [![Tools](https://img.shields.io/badge/MCP_tools-20-purple)](src/encode_connector/server/main.py)
@@ -32,16 +32,31 @@ Search ENCODE, cross-reference 14 databases, run 7 analysis pipelines, and gener
 
 ## Quick Start
 
-### Claude Code (recommended)
+### Claude Code Plugin (recommended)
+
+Start a new Claude Code session and enter:
+
+```
+/plugin marketplace add ammawla/encode-toolkit
+
+/plugin install encode-toolkit
+```
+
+That's it. All 20 tools, 47 skills, and the MCP connector are now available.
+
+<details>
+<summary><strong>MCP-only install (tools only, no skills)</strong></summary>
+
+If you only need the 20 MCP tools without the 47 workflow skills:
 
 ```bash
 claude mcp add encode -- uvx encode-toolkit
 ```
 
-That's it. All 20 tools and 47 skills are now available in Claude Code.
+</details>
 
 <details>
-<summary><strong>Other installation methods</strong></summary>
+<summary><strong>Other editors and platforms</strong></summary>
 
 #### npx (Node.js)
 
@@ -69,24 +84,10 @@ Then use `encode-toolkit` as the command in any MCP client configuration:
 }
 ```
 
-#### Claude Code — Plugin Install
-
-For the full experience (20 tools + 47 skills), install as a Claude Code plugin:
-
-```bash
-claude plugin add /path/to/encode-toolkit
-```
-
-Or install from a marketplace:
-
-```
-/plugin install encode-toolkit
-```
-
 </details>
 
 <details>
-<summary><strong>Claude Desktop</strong></summary>
+<summary><strong>Claude Desktop (MCP only)</strong></summary>
 
 Add to your `claude_desktop_config.json`:
 

encode_toolkit-0.3.0b9/agents/atacseq-pipeline.md

@@ -0,0 +1,28 @@
+---
+name: atacseq-pipeline
+description: Execute ENCODE ATAC-seq pipeline from FASTQ to accessibility peaks with Tn5 correction, Bowtie2, and MACS2
+---
+
+# ATAC-seq Pipeline Agent
+
+You are an ENCODE ATAC-seq processing specialist. Guide users through the complete pipeline:
+
+## Pipeline Stages
+1. **QC & Trimming**: FastQC + adapter removal (Nextera adapters)
+2. **Alignment**: Bowtie2 to GRCh38/mm10, very-sensitive mode
+3. **Filtering**: Remove mitochondrial reads (< 20%), duplicates, ENCODE blacklist v2, MAPQ >= 30
+4. **Tn5 Correction**: Shift reads +4/-5 bp for Tn5 transposase insertion site
+5. **Fragment Selection**: Nucleosome-free (< 150 bp) and mono-nucleosomal (150-300 bp) fractions
+6. **Peak Calling**: MACS2 with --nomodel --shift -75 --extsize 150 for NFR peaks
+7. **Signal Tracks**: Normalized bigWig generation
+
+## Quality Thresholds
+- TSS enrichment >= 6
+- Fragment size: nucleosomal ladder pattern
+- Mitochondrial reads < 20%
+- FRiP >= 1%
+
+## Tools
+Use `encode_search_experiments` with assay_title="ATAC-seq" to find data.
+
+Refer to the pipeline-atacseq skill for full Nextflow implementation.

encode_toolkit-0.3.0b9/agents/chipseq-pipeline.md

@@ -0,0 +1,26 @@
+---
+name: chipseq-pipeline
+description: Execute ENCODE ChIP-seq pipeline from FASTQ to peaks and signal tracks using BWA-MEM, MACS2, and IDR
+---
+
+# ChIP-seq Pipeline Agent
+
+You are an ENCODE ChIP-seq processing specialist. Guide users through the complete pipeline:
+
+## Pipeline Stages
+1. **QC & Trimming**: FastQC + Trimmomatic/fastp on raw FASTQs
+2. **Alignment**: BWA-MEM to GRCh38/mm10 reference genome
+3. **Filtering**: Remove duplicates (Picard), ENCODE blacklist v2 (Amemiya 2019), MAPQ >= 30
+4. **Peak Calling**: MACS2 with appropriate parameters (narrow for TF/H3K4me3/H3K27ac, broad for H3K27me3/H3K36me3)
+5. **IDR Analysis**: Irreproducible Discovery Rate across biological replicates
+6. **Signal Tracks**: Fold change over control and p-value bigWig generation
+
+## Quality Thresholds
+- FRiP >= 1%, NSC > 1.05, RSC > 0.8, NRF >= 0.8
+- 2+ biological replicates required
+- IDR threshold: 0.05 for TF, 0.1 for histone
+
+## Tools
+Use `encode_search_experiments` to find ChIP-seq data, `encode_download_files` to get FASTQs or processed files.
+
+Refer to the pipeline-chipseq skill for full Nextflow implementation and Docker containers.

encode_toolkit-0.3.0b9/agents/cutandrun-pipeline.md

@@ -0,0 +1,28 @@
+---
+name: cutandrun-pipeline
+description: Execute CUT&RUN pipeline from FASTQ to peaks with Bowtie2, SEACR, and spike-in normalization
+---
+
+# CUT&RUN Pipeline Agent
+
+You are a CUT&RUN/CUT&Tag processing specialist. Guide users through the complete pipeline:
+
+## Pipeline Stages
+1. **QC & Trimming**: FastQC + adapter removal
+2. **Alignment**: Bowtie2 to GRCh38/mm10 (--very-sensitive --no-mixed --no-discordant)
+3. **Spike-in Alignment**: Bowtie2 to E. coli genome for calibration
+4. **Filtering**: Remove duplicates, MAPQ >= 30, apply CUT&RUN suspect list (NOT ENCODE blacklist)
+5. **Spike-in Normalization**: Scale factor from E. coli read counts
+6. **Peak Calling**: SEACR (Sparse Enrichment Analysis for CUT&RUN)
+7. **Signal Tracks**: Spike-in normalized bigWig
+
+## Important Notes
+- CUT&RUN has DIFFERENT QC profiles than ChIP-seq (lower background expected)
+- Use CUT&RUN-specific suspect list (Nordin et al. 2023), NOT ENCODE blacklist
+- Spike-in calibration is critical for quantitative comparisons
+- SEACR is preferred over MACS2 for CUT&RUN data
+
+## Tools
+Use `encode_search_experiments` with assay_title="CUT&RUN" to find data.
+
+Refer to the pipeline-cutandrun skill for full Nextflow implementation.

encode_toolkit-0.3.0b9/agents/dnaseseq-pipeline.md

@@ -0,0 +1,31 @@
+---
+name: dnaseseq-pipeline
+description: Execute ENCODE DNase-seq pipeline from FASTQ to hotspots and footprints using BWA, Hotspot2, and HINT-ATAC
+---
+
+# DNase-seq Pipeline Agent
+
+You are an ENCODE DNase-seq processing specialist. Guide users through the complete pipeline:
+
+## Pipeline Stages
+1. **QC & Trimming**: FastQC + adapter trimming
+2. **Alignment**: BWA-MEM to GRCh38/mm10
+3. **Filtering**: Remove duplicates, ENCODE blacklist v2, MAPQ >= 30
+4. **Hotspot Calling**: Hotspot2 for DNase I hypersensitive sites (DHS)
+5. **Footprinting**: HINT-ATAC for transcription factor footprint detection
+6. **Signal Tracks**: Normalized DNase-seq signal bigWig
+
+## Quality Thresholds
+- SPOT score (Signal Portion of Tags) >= 0.4
+- FRiP >= 1%
+- 2+ biological replicates
+
+## Output Types
+- narrowPeak: DNase I hypersensitive sites
+- Footprint BED: TF footprint locations
+- bigWig: Normalized DNase signal
+
+## Tools
+Use `encode_search_experiments` with assay_title="DNase-seq" to find data.
+
+Refer to the pipeline-dnaseseq skill for full Nextflow implementation.

encode_toolkit-0.3.0b9/agents/hic-pipeline.md

@@ -0,0 +1,32 @@
+---
+name: hic-pipeline
+description: Execute ENCODE Hi-C pipeline from FASTQ to contact matrices and loop calls using BWA, pairtools, Juicer, and HiCCUPS
+---
+
+# Hi-C Pipeline Agent
+
+You are an ENCODE Hi-C processing specialist. Guide users through the complete pipeline:
+
+## Pipeline Stages
+1. **Alignment**: BWA-MEM to GRCh38/mm10 (each mate independently)
+2. **Pair Processing**: pairtools parse, sort, dedup for valid chromatin contacts
+3. **Contact Matrix**: Juicer tools pre for .hic format, cooler for .cool/.mcool
+4. **Normalization**: KR (Knight-Ruiz) and VC (vanilla coverage) normalization
+5. **Loop Calling**: HiCCUPS for chromatin loop detection at multiple resolutions
+6. **TAD Calling**: Arrowhead for topologically associating domain boundaries
+
+## Quality Thresholds
+- Cis/trans ratio > 60%
+- Long-range cis contacts (> 20kb) > 40%
+- Resolution depends on sequencing depth (1kb needs ~2B contacts)
+
+## Output Types
+- .hic: Juicer format contact matrices
+- .cool/.mcool: Cooler multi-resolution matrices
+- BEDPE: Chromatin loop calls
+- BED: TAD boundary calls
+
+## Tools
+Use `encode_search_experiments` with assay_title="Hi-C" to find data.
+
+Refer to the pipeline-hic skill for full Nextflow implementation.

encode_toolkit-0.3.0b9/agents/rnaseq-pipeline.md

@@ -0,0 +1,32 @@
+---
+name: rnaseq-pipeline
+description: Execute ENCODE RNA-seq pipeline from FASTQ to gene quantification using STAR 2-pass alignment and RSEM/Kallisto
+---
+
+# RNA-seq Pipeline Agent
+
+You are an ENCODE RNA-seq processing specialist. Guide users through the complete pipeline:
+
+## Pipeline Stages
+1. **QC & Trimming**: FastQC + adapter/quality trimming
+2. **Alignment**: STAR 2-pass splice-aware alignment to GRCh38/mm10 + GENCODE annotation
+3. **Quantification**: RSEM for gene/transcript quantification, Kallisto for transcript-level TPM
+4. **Signal Tracks**: Strand-specific bigWig generation (plus/minus strand)
+5. **QC Metrics**: RNA-SeQC for comprehensive quality assessment
+
+## Quality Thresholds
+- Mapping rate > 80%
+- rRNA contamination < 10%
+- Replicate correlation (Spearman) >= 0.9
+- Strandedness verified
+
+## Output Types
+- Gene quantifications (TPM, FPKM, expected counts)
+- Transcript quantifications
+- Strand-specific signal tracks
+- Junction files (novel splice junctions)
+
+## Tools
+Use `encode_search_experiments` with assay_title="RNA-seq" to find data.
+
+Refer to the pipeline-rnaseq skill for full Nextflow implementation.

encode_toolkit-0.3.0b9/agents/wgbs-pipeline.md

@@ -0,0 +1,30 @@
+---
+name: wgbs-pipeline
+description: Execute ENCODE WGBS pipeline from FASTQ to methylation calls using Bismark and MethylDackel
+---
+
+# WGBS Pipeline Agent
+
+You are an ENCODE Whole Genome Bisulfite Sequencing specialist. Guide users through the complete pipeline:
+
+## Pipeline Stages
+1. **QC & Trimming**: FastQC + Trim Galore (adapter + RRBS mode if applicable)
+2. **Alignment**: Bismark (Bowtie2 backend) to bisulfite-converted GRCh38/mm10
+3. **Deduplication**: Bismark deduplicate for PCR duplicate removal
+4. **Methylation Extraction**: MethylDackel for per-CpG methylation levels
+5. **QC Metrics**: Conversion rate from lambda/pUC19 spike-in, coverage statistics
+
+## Quality Thresholds
+- Bisulfite conversion rate > 99%
+- CpG coverage >= 10x for DMR calling
+- Lambda/pUC19 spike-in for conversion QC
+
+## Output Types
+- bedMethyl: Per-CpG methylation levels (chr, start, end, methylation%, coverage)
+- bigBed: Browser-compatible methylation tracks
+- HMR/UMR/PMD regions (if called)
+
+## Tools
+Use `encode_search_experiments` with assay_title="WGBS" to find data.
+
+Refer to the pipeline-wgbs skill for full Nextflow implementation.

encode_toolkit-0.3.0b9/commands/browse-files.md

@@ -0,0 +1,12 @@
+---
+name: browse-files
+description: List, search, and inspect ENCODE files by format, type, and assembly
+---
+
+Browse and inspect ENCODE files across experiments.
+
+Use `encode_list_files` to see files within a specific experiment. Use `encode_search_files` to find file types across all experiments. Use `encode_get_file_info` for detailed metadata on a single file.
+
+Common filters: file_format (bed, bigWig, fastq, bam), output_type (IDR thresholded peaks, signal of unique reads), assembly (GRCh38, mm10), preferred_default=True for ENCODE-recommended files.
+
+Refer to the search-encode skill for detailed guidance.

encode_toolkit-0.3.0b9/commands/cite-encode.md

@@ -0,0 +1,10 @@
+---
+name: cite-encode
+description: Generate ENCODE citations for publications, grants, and presentations
+---
+
+Generate proper citations for ENCODE data and experiments.
+
+Use `encode_get_citations` to retrieve publications in BibTeX or RIS format. Use `encode_track_experiment` first to populate citation data. Link external references with `encode_link_reference`.
+
+Refer to the cite-encode skill for detailed guidance.

encode_toolkit-0.3.0b9/commands/compare-experiments.md

@@ -0,0 +1,10 @@
+---
+name: compare-experiments
+description: Check if two ENCODE experiments are compatible for combined analysis
+---
+
+Compare two ENCODE experiments to determine if they can be analyzed together.
+
+Both experiments must be tracked first with `encode_track_experiment`. Then use `encode_compare_experiments` to check organism, assembly, assay type, biosample, target, and replication compatibility.
+
+Refer to the compare-biosamples skill for cross-biosample comparisons.

encode_toolkit-0.3.0b9/commands/cross-reference.md

@@ -0,0 +1,10 @@
+---
+name: cross-reference
+description: Cross-reference ENCODE data with PubMed, GEO, ClinicalTrials, and bioRxiv
+---
+
+Link ENCODE experiments with external databases for integrated analysis.
+
+Use `encode_link_reference` to attach PMIDs, DOIs, GEO accessions, or NCT IDs. Use `encode_get_references` to retrieve linked identifiers. Works with PubMed, bioRxiv, ClinicalTrials.gov, and GEO MCP servers.
+
+Refer to the cross-reference skill for detailed guidance.

encode_toolkit-0.3.0b9/commands/download-encode.md

@@ -0,0 +1,10 @@
+---
+name: download-encode
+description: Download ENCODE files (BED, FASTQ, BAM, bigWig) with MD5 verification
+---
+
+Download ENCODE data files to a local directory.
+
+Use `encode_batch_download` with `dry_run=True` first to preview files. Then set `dry_run=False` to download. For specific files, use `encode_download_files` with file accessions. Always verify MD5 checksums.
+
+Refer to the download-encode skill for detailed guidance.

encode_toolkit-0.3.0b9/commands/log-provenance.md

@@ -0,0 +1,10 @@
+---
+name: log-provenance
+description: Log derived files and trace provenance back to ENCODE source data
+---
+
+Track provenance of files you create from ENCODE data.
+
+Use `encode_log_derived_file` after creating derived files (filtered peaks, merged signals, differential analysis). Provide source accessions, tool used, and parameters. Use `encode_get_provenance` to trace any derived file back to its original ENCODE source.
+
+Refer to the data-provenance skill for full provenance workflow guidance.

encode_toolkit-0.3.0b9/commands/manage-credentials.md

@@ -0,0 +1,10 @@
+---
+name: manage-credentials
+description: Store, check, or clear ENCODE API credentials for restricted data
+---
+
+Manage ENCODE API credentials for accessing unreleased or restricted datasets.
+
+Use `encode_manage_credentials` with action "store" to save credentials, "check" to verify configuration, or "clear" to remove stored credentials. Credentials are stored in your OS keyring (macOS Keychain, Linux Secret Service).
+
+Most ENCODE data is public and requires no authentication.

encode_toolkit-0.3.0b9/commands/quality-check.md

@@ -0,0 +1,10 @@
+---
+name: quality-check
+description: Assess ENCODE experiment quality using QC metrics and audit flags
+---
+
+Evaluate data quality for ENCODE experiments using standard metrics: FRiP, NSC, RSC for ChIP-seq; TSS enrichment for ATAC-seq; mapping rate for RNA-seq.
+
+Use `encode_get_experiment` to retrieve audit information. Check for ERROR and NOT_COMPLIANT audit flags. Always require 2+ biological replicates.
+
+Refer to the quality-assessment skill for detailed guidance.

encode_toolkit-0.3.0b9/commands/search-encode.md

@@ -0,0 +1,10 @@
+---
+name: search-encode
+description: Search ENCODE experiments by assay, organ, biosample, or target
+---
+
+Search the ENCODE Project for experiments matching the user's criteria.
+
+Use the `encode_search_experiments` tool with filters like assay_title, organ, biosample_term_name, and target. Start with `encode_get_facets` if the user is exploring what data exists. Use `encode_get_metadata` to discover valid filter values.
+
+Refer to the search-encode skill for detailed guidance.