SkillAgentSearch skills...

pdb-database

Use when you want to search for or download experimentally-determined 3D structures for biomolecules (proteins, nucleic acids, bound ligands). Supports searching by sequence similarity, structure similarity, chemical and other attributes.

Install / Use

npx skills add google-deepmind/science-skills --skill pdb_database

Installs into whichever agent you are using.

About this skill
📄

SKILL.md

Installable skill definition

Quality Score

93/100

Supported Platforms

Universal

Our assessment of pdb-database

pdb-database scores 93/100 on our quality scale, 119th of 394 Communication skills we index (top 31%).

Its SKILL.md is 8.7 KB long, well organised into 21 sections with 12 code examples: a thorough specification that gives an agent plenty to work with.

With 3,137 GitHub stars, it is one of the more widely adopted skills in the catalogue.

Substance
29/30
Structure
20/20
Description
15/15
Adoption
15/20
Freshness
15/15

Maintenance, license and trust

  • The repository was last updated about 2 months ago, so pdb-database is actively maintained.
  • Our last check on 2026-10-26 found the source still online.
  • It is released under the Apache-2.0 license, a permissive license that allows use, modification and commercial use with attribution.
  • Its trust signals score 100/100, with no cautions. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.

Safety scan

No issues found

Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful.

AI review by kimi-k2.7-code on 2026-09-29. Automated pattern scan on 2026-09-29. It catches known dangerous patterns, not every risk — read a skill before letting an agent act on it.

pdb-database compared with similar skills

All 4 of these similar skills score higher than pdb-database; compare them before choosing.

SkillScoreStarsUpdatedFormat
pdb-database (this skill)by google-deepmind933.1k2mo agoSKILL.md
x402by browser-use100121.0k3d agoSKILL.md
algorithmic-artby anthropics100177.9k55d agoSKILL.md
pptxby anthropics100177.9k55d agoSKILL.md
designby nextlevelbuilder100149.0k6d agoSKILL.md

Frequently asked questions

How do I install pdb-database?
Run npx skills add google-deepmind/science-skills --skill pdb-database. The install tabs above show the steps for each supported agent.
Which AI agents does pdb-database work with?
It is written for Universal, as a SKILL.md file. Other agents that read the same format can often use it too.
Is pdb-database safe to use?
Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful. It is Apache-2.0-licensed and scores 100/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
Is pdb-database still maintained?
The repository was last updated about 2 months ago, so pdb-database is actively maintained.

name: pdb-database description: > Use when you want to search for or download experimentally-determined 3D structures for biomolecules (proteins, nucleic acids, bound ligands). Supports searching by sequence similarity, structure similarity, chemical and other attributes. Also use to get metadata about biomolecular structure experiments.

RCSB Protein Data Bank skill

Prerequisites

  1. uv: Read the uv skill and follow its Setup instructions to ensure uv is installed and on PATH.
  2. User Notification: If .licenses/pdb_database_LICENSE.txt does not already exist in the workspace root directory then (1) prominently notify the user to check the terms at https://www.rcsb.org/pages/usage-policy, then (2) create the file recording the notification text and timestamp.

Core Rules

  • Always prefer to use the provided scripts. Only as a last resort use curl, urllib, raw HTTP requests, or any other method to access PDB APIs. The scripts automatically enforce required rate limits.
  • Always redirect output to a file. Parse output with e.g. jq, grep, or a short Python snippet. Do NOT print large API responses to stdout to avoid truncation.
  • Notification: If this skill is used, ensure this is mentioned in the output.
  • Explain your queries On completing a task that used PDB JSON/GraphQL queries, explain in clear language what your query did so the user can correct any bad assumptions.

Attribute-based search workflow

  1. Fetch the relevant schema to discover searchable attribute names. For structure attributes: uv run scripts/fetch_schema.py --api search_structure --output schema_structure.txt For chemical attributes: uv run scripts/fetch_schema.py --api search_chemical --output schema_chemical.txt

  2. Grep the schema to find relevant attributes. Grep one keyword at a time and examine many lines — there are lots of similar attributes and you must choose the best match for the user's intent.

  3. Compose and run a JSON search query using the discovered attributes: uv run scripts/search_pdb.py --query '<JSON>' --return_type <RETURN_TYPE> --output results.json Pass the --count_only flag to get just the number of matching entries.

For step 2: some basic PDB concepts (helpful for attribute choice)

  • Entity: A unique molecule found in a structure.
  • Instance / Chain: A particular copy of an entity. E.g. if a structure contains two protein chains with the same sequence, they are the same entity but different instances / chains.
  • Assembly: A biologically relevant collection of instances / chains. This may be the same as the deposited structure, a subset, or multiple copies.
  • Label vs Auth: Polymer instances get letter labels ("A", "B", "AA") and their monomers are numbered. There are author-assigned ("auth") and PDB-internal ("label") schemes. The label scheme is more consistent and is always used in scripts and APIs. However, users and papers may refer to the author scheme (clarify which scheme is being used if necessary).
  • Chemical component: A small molecule / monomer, with an ID matching [A-Z]{1,3}
  • Primary citation: The main publication about a structure. Prefer primary_citation attributes over citation attributes.
  • Resolution: Frequently used measure of structure quality (lower is better). Usually prefer rcsb_entry_info.resolution_combined, which accounts for different experimental methods.

For step 3: Example queries

# Non-human proteins published in Nature, newest first
uv run scripts/search_pdb.py --query '{ "type": "group", "logical_operator": "and", "nodes": [ { "type": "terminal", "service": "text", "parameters": { "operator": "exact_match", "negation": true, "value": "Homo sapiens", "attribute": "rcsb_entity_source_organism.taxonomy_lineage.name" } }, { "type": "terminal", "service": "text", "parameters": { "operator": "exact_match", "value": "Nature", "attribute": "rcsb_primary_citation.rcsb_journal_abbrev" } } ] }' --return_type entry --sort_by rcsb_accession_info.initial_release_date --sort_direction desc --page_start 0 --rows 100 --output results.json
# Structures containing the chemical component CA (Ca2+ ion)
uv run scripts/search_pdb.py --query '{ "type": "terminal", "service": "text_chem", "parameters": { "operator": "exact_match", "value": "CA", "attribute": "rcsb_chem_comp_container_identifiers.comp_id" } }' --return_type entry --output results.json
# Number of entries with disulfide bonds
uv run scripts/search_pdb.py --query '{ "type": "terminal", "service": "text", "parameters": { "operator": "exact_match", "value": "disulfide bridge", "attribute": "rcsb_polymer_struct_conn.connect_type" } }' --return_type entry --count-only --output count.json

Common operators: exact_match, equals, exists, contains_phrase, contains_words, in, greater, less

Similarity-based search workflow

Similarity searches do not require a schema fetch. Basic examples:

# Sequence similarity
uv run scripts/search_pdb.py --query '{ "query": { "type": "terminal", "service": "sequence", "parameters": { "evalue_cutoff": 1, "identity_cutoff": 0.9, "sequence_type": "protein", "value": "MTEYKLVVVGAGGVGKSALTIQLIQNHFVDEYDPTIEDSYRKQ" } }, "request_options": { "scoring_strategy": "sequence" } }' --return_type polymer_entity --output results.json
# Structure similarity
uv run scripts/search_pdb.py --query '{ "type": "terminal", "service": "structure", "parameters": { "value": {"entry_id": "6LU7", "asym_id": "A"}, "number_of_candidates": 2000 } }' --return_type polymer_entity --output results.json
# Sequence motif match
uv run scripts/search_pdb.py --query '{ "type": "terminal", "service": "seqmotif", "parameters": { "value": "C-x(2,4)-C-x(3)-[LIVMFYWC]-x(8)-H-x(3,5)-H.", "pattern_type": "prosite", "sequence_type": "protein" } }' --return_type polymer_entity --output results.json
# Chemical descriptor match
uv run scripts/search_pdb.py --query '{ "type": "terminal", "service": "chemical", "parameters": { "value": "InChI=1S/C8H9NO2/c1-6(10)9-7-2-4-8(11)5-3-7/h2-5,11H,1H3,(H,9,10)", "type": "descriptor", "descriptor_type": "InChI", "match_type": "graph-strict" } }' --return_type mol_definition --output results.json

See https://search.rcsb.org/#search-services for more details.

Full text search workflow

Searches all text associated with an entry. Example:

uv run scripts/search_pdb.py --query '{ "type": "terminal", "service": "full_text", "parameters": { "value": "isopeptide + ( collagen | fibrinogen )" } }' --return_type entry --output results.json

Important: use full_text search as a last resort when there's no more precise attribute search available. Consider using the struct.title or rcsb_pubmed_abstract_text attributes instead.

File download workflow

To download full PDB entries, use the download_coordinate_files.py script. Use this when you need access to atomic coordinates, when asked for a pdb / mmcif file, or when non-specifically asked to fetch a PDB code. Example:

uv run scripts/download_coordinate_files.py --ids "4HHB,6BEA" --format "mmcif" --output_dir <OUTPUT_DIR>

Metadata query workflow

This flow is significantly more efficient than downloading full coordinate files when you only need a few pieces of metadata about each entry / entity.

  1. Fetch the schema for the relevant object type. E.g. uv run scripts/fetch_schema.py --api data_entry --output schema_entry.txt

  2. Grep the schema for relevant fields (one keyword at a time, many lines).

  3. Compose and run a GraphQL metadata query: uv run scripts/fetch_pdb_metadata.py --query '<GraphQL>' --output results.json

For step 3: Example queries

# Fetch structure titles and experimental methods
uv run scripts/fetch_pdb_metadata.py --query '{ entries(entry_ids: ["1STP", "2JEF", "1CDG"]) { rcsb_id struct { title } exptl { method } } }' --output results.json
# Fetch polymer entity taxonomy and cluster membership
uv run scripts/fetch_pdb_metadata.py --query '{ polymer_entities(entity_ids:["2CPK_1","3WHM_1","2D5Z_1"]) { rcsb_id rcsb_entity_source_organism { ncbi_taxonomy_id ncbi_scientific_name } rcsb_cluster_membership { cluster_id identity } } }' --output results.json
# Fetch polymer entity external sequence database accessions
uv run scripts/fetch_pdb_metadata.py --query '{ entries(entry_ids:["7NHM", "5L2G"]){ polymer_entities { rcsb_id rcsb_polymer_entity_container_identifiers { reference_sequence_identifiers { database_accession database_name } } } } }' --output results.json

Related Skills

View on GitHub
GitHub Stars3.1k
CategoryCommunication
Updated2mo ago
Forks351

Languages

Python

Trust signals

100/100

From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.

No cautions