interpro-database
Identify domains, families, and sites in proteins; find all proteins in a family or sharing a domain; explore species distribution for a domain; annotate genomes with protein families and GO terms. InterPro combines 14 databases (e.g., Pfam, CDD) into one searchable resource.
Install / Use
npx skills add google-deepmind/science-skills --skill interpro_databaseInstalls into whichever agent you are using.
SKILL.md
Installable skill definition
Quality Score
Category
Data & AnalyticsSupported Platforms
Our assessment of interpro-database
interpro-database scores 95/100 on our quality scale, 76th of 604 Data & Analytics skills we index (top 13%).
Its SKILL.md is 17 KB long, well organised into 24 sections with 8 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.
Maintenance, license and trust
- The repository was last updated about 2 months ago, so interpro-database is actively maintained.
- Our last check on 2026-11-16 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 foundOur 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.
interpro-database compared with similar skills
All 4 of these similar skills score higher than interpro-database; compare them before choosing.
| Skill | Score | Stars | Updated | Format |
|---|---|---|---|---|
| interpro-database (this skill)by google-deepmind | 95 | 3.1k | 2mo ago | SKILL.md |
| x402by browser-use | 100 | 121.0k | 3d ago | SKILL.md |
| algorithmic-artby anthropics | 100 | 177.9k | 55d ago | SKILL.md |
| pptxby anthropics | 100 | 177.9k | 55d ago | SKILL.md |
| designby nextlevelbuilder | 100 | 149.0k | 6d ago | SKILL.md |
Frequently asked questions
- How do I install interpro-database?
- Run
npx skills add google-deepmind/science-skills --skill interpro-database. The install tabs above show the steps for each supported agent. - Which AI agents does interpro-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 interpro-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 interpro-database still maintained?
- The repository was last updated about 2 months ago, so interpro-database is actively maintained.
Skill content
View source on GitHubname: interpro-database description: > Identify domains, families, and sites in proteins; find all proteins in a family or sharing a domain; explore species distribution for a domain; annotate genomes with protein families and GO terms. InterPro combines 14 databases (e.g., Pfam, CDD) into one searchable resource. InterPro-N significantly expands annotation and sequence coverage with deep learning. Includes domain architecture (IDA) search.
InterPro Database Access
Prerequisites
uv: Read theuvskill and follow its Setup instructions to ensureuvis installed and on PATH.- User Notification: If .licenses/interpro_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.ebi.ac.uk/interpro/ and https://www.ebi.ac.uk/about/terms-of-use/, then (2) create the file recording the notification text and timestamp.
Overview
InterPro combines signatures from multiple, diverse databases into a single searchable resource, reducing redundancy and helping users interpret their sequence analysis results. By uniting these member databases (e.g., Pfam, CDD, SMART), InterPro capitalises on their individual strengths to produce a powerful diagnostic tool and integrated resource.
Use interpro-database to:
- Identify what domains, families, and sites are found in a particular protein.
- Identify all proteins that belong to a protein family or contain a particular domain, even when the names and activities of the proteins are highly variable.
- Examine the species in which a particular protein family or domain is found.
- Annotate genomes with protein family information and Gene Ontology (GO) terms.
This skill provides a robust utility, interpro_client.py, to interact with the
InterPro API seamlessly. It natively handles rate limiting (HTTP 429),
background query sleep tracking (HTTP 408), terminal errors (HTTP 404/410), and
lazy pagination.
Core Rules
- Use the Wrapper: ALWAYS execute the
scripts/interpro_client.pyhelper script to query the database rather than accessing the database directly. The scripts automatically enforce fair use and implement retry logic. - For exploratory queries: ALWAYS use the CLI with a strict
--limit. This allows you to rapidly understand the data schema without polluting your context window or fetching millions of results. - Output to file: Use the CLI with --output to output to a file rather than attempting to print it all to the console. Process the output using jq or code.
- For more complex pipelines import the module natively into your Python scripts to consume the generator directly, preventing the need to deserialize CLI strings in large workflows.
- Notification: If this skill is used, ensure this is mentioned in the output.
Examples:
uv run ./scripts/interpro_client.py fetch protein --source_db reviewed --limit 2 --query_params tax_id=9606 --output exploratory_results.jsonl
import sys
sys.path.append('scripts')
from interpro_client import fetch_interpro_data
import itertools
# fetch_interpro_data lazily yields results page-by-page
results = fetch_interpro_data(
endpoint="entry",
source_db="pfam",
query_params={"page_size": 10}
)
for match in itertools.islice(results, 10):
print(match["metadata"]["accession"])
4 Ways to Construct Endpoints:
The arguments strictly map to the four common API path constructions. Do not
format your own / separated strings:
/{endpoint}(e.g./entry)uv run ./scripts/interpro_client.py fetch entry --limit 10 --output entries.jsonl/{endpoint}/{sourceDB}(e.g./entry/pfam)uv run ./scripts/interpro_client.py fetch entry --source_db pfam --limit 10 --output pfam_entries.jsonl/{endpoint}/{sourceDB}/{accession}(e.g./entry/pfam/PF00001)uv run ./scripts/interpro_client.py fetch entry --source_db pfam --accession PF00001 --limit 10 --output pf00001_entry.jsonl/{endpoint}/{sourceDB}/{linked_endpoint}/{sourceDB}/{accession}(e.g./entry/interpro/protein/uniprot/P04637)uv run ./scripts/interpro_client.py fetch entry \ --source_db interpro \ --linked_endpoint protein \ --linked_source_db uniprot \ --linked_accession P04637 \ --limit 10 --output p04637_entries.jsonl
Valid Source Databases (--source_db)
Each endpoint only accepts specific source_db values. Using an invalid value
returns a 404 error.
/entry(16 values):interpro,pfam,cathgene3d,ssf,panther,cdd,profile,smart,ncbifam,prosite,prints,hamap,pirsf,sfld,antifam./protein(3 values):uniprot(all),reviewed(SwissProt),unreviewed(TrEMBL)./structure(1 value):pdb./taxonomy(1 value):uniprot./proteome(1 value):uniprot./set(2 values):pfam,cdd.
Quick Reference / Core Endpoints & Parameters
For a complete, exhaustive list of all query parameters, see the Full API Reference.
The API is fully open and supports 6 core endpoints. You can combine them using the linked parameters described above. Below is a nested list of the specific query parameters available for each endpoint:
-
/entry(Domain, family, active site, repeat, or homologous superfamily entries)integrated: Filter by integrated status (e.g.,pfam).type: Filter by type (e.g.,family,domain,homologous_superfamily).go_term/go_category: Filter by Gene Ontology.ida_search/ida_ignore/exact/ordered: Filter by domain architecture (see IDA Search section).extra_fields: Request additional data (e.g.,countersfor match coordinates).group_by/sort_by: Aggregate or sort results (valid values depend on context, see Full API Reference).- Example:
uv run ./scripts/interpro_client.py count entry --source_db pfam --query_params type=domain --output count.jsonl
-
/protein(Protein records matching entries or domains)tax_id: Filter by taxonomy ID (does not search lineage).match_presence: Filter by proteins having InterPro matches (true/false).is_fragment: Filter complete vs. fragment sequences.group_by: Aggregate results (e.g.,taxonomy).extra_fields: Request sequence or match details.isoforms/residues/structureinfo: Include specific sub-features.conservation/extra_features: Append residue conservation flags or Mobidb/coil features (only valid for/protein/{source_db}/{accession}).- Example:
uv run ./scripts/interpro_client.py fetch protein --source_db uniprot --limit 20 --query_params tax_id=9606 --output human_proteins.jsonl
-
/structure(PDB structures linked to InterPro entries)experiment_type: Filter by experimental method (e.g.,X-RAY DIFFRACTION).resolution: Filter by resolution limit.extra_fields: Include additional structural metadata.group_by: Aggregate results.- Example:
./scripts/interpro_client.py fetch structure --source_db pdb --accession 1ATP --limit 10 --output 1atp_structures.jsonl
-
/taxonomy(Taxonomy distribution nodes)key_species: Filter to limit to key species.with_names: Include scientific names.filter_by_entry/filter_by_entry_db: Filter intersection with specific entries.extra_fields: Additional taxonomic metadata.- Example:
./scripts/interpro_client.py fetch taxonomy --source_db uniprot --accession 9606 --limit 10 --output human_taxonomy.jsonl
-
/proteome(Complete proteomes linked to InterPro)extra_fields: General query expansion.- Example:
uv run ./scripts/interpro_client.py fetch proteome --source_db uniprot --accession UP000005640 --limit 10 --output proteome.jsonl
-
/set(Curated sets of related entries, e.g., Pfam clans)extra_fields: Additional metadata (only valid for/set/{sourceDB}).- Example:
uv run ./scripts/interpro_client.py fetch set --source_db pfam --accession CL0001 --limit 10 --output pfam_clan.jsonl
InterPro Domain Architecture (IDA) Search
InterPro provides powerful tools for searching proteins by their domain architecture (the exact combination and order of domains). Because the API does not allow querying proteins directly by multiple domains at once (e.g., "give me proteins with PF00069 AND PF00017"), finding proteins with specific domain combinations requires a two-step process.
Step 1: Find matching architectures (ida_search)
The ida_search parameter is used on the root /entry endpoint to find all
Domain Architectures (IDAs) containing the domains you specify.
- Constraints:
- Valid ONLY on the root
/entryendpoint. - Cannot be combined with non-IDA parameters.
- Valid ONLY on the root
- Modifiers (Only valid with
ida_search):ida_ignore: Ignores the given domains in the search (query param).ordered: Ensures domains appear in the exact specified order (flag).exact: Ensures the architecture matches exactly (no additional domains) (flag). Requiresorderedflag to be present.
Example: Find architectures containing both a kinase domain (PF00069) and an SH2 domain (PF00017), in that exact order:
uv run scripts/interpro_client.py fetch entry
--query_params ida_search=PF00069,PF00017
--flags ordered exact
--output architectures.jsonl
Note: This returns the architectures and their unique ida_ids, not all
individual proteins.
Step 2: Fetch proteins for those architectures (ida)
Once you have the ida_ids (e.g., 619edbb...) from Step 1, you can fetch all
the actual proteins that share that precise layout by filtering the /protein
endpoint.
Constraints:
- Valid on
/proteinand/entry/{sourceDB}/{accession}endpoints.
Example: Fetch proteins matching one of the architecture IDs from Step 1:
uv run scripts/interpro_client.py fetch protein
--source_db uniprot
--query_params ida=619edbb2b445bfa3ad51bd894e3c115b025a5f25
--output matching_proteins.jsonl
(When building pipelines or querying comprehensively, you would loop through
all the ida_ids from Step 1 and run Step 2 for each one).
InterPro Entry Types
Each InterPro entry is assigned a type indicating what you can infer when a protein matches the entry:
- Domain: Distinct functional, structural or sequence units that may exist in a variety of biological contexts. Example: PH domain or classical C2H2 zinc finger.
- Family: A group of proteins sharing a common evolutionary origin reflected by related functions, sequence similarities, or primary/secondary/tertiary structures.
- Homologous Superfamily: Proteins sharing an evolutionary origin reflected by structural similarity but often displaying very low sequence similarity. Usually comprises signatures from the SUPERFAMILY and CATH-Gene3D databases.
- Repeat: A short sequence that is typically repeated within a protein, often <50 amino acids long. Example: Leucine Rich Repeats or WD40 repeats.
- Site: Includes
Active site(sequence containing conserved residues for catalytic activity) andBinding site(sequence containing conserved residues forming a protein interaction site).
InterPro-N Predictions
InterPro-N is a deep-learning-based exte
Truncated for display — read the full file on GitHub.
Related Skills
x402
121.0kSet up Browser Use Cloud payments with x402 — pay per request from a crypto wallet (USDC on Base mainnet), no signup or API key.
algorithmic-art
177.9kCreating algorithmic art using p5.js with seeded randomness and interactive parameter exploration. Use this when users request creating art using code, generative art, algorithmic art, flow fields, or particle systems.
pptx
177.9kUse this skill any time a .pptx or .potx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx or .potx file (even if the extracted content will be used elsewhere, like in an em…
design
149.0kComprehensive design skill: brand identity, design tokens, UI styling, logo generation (55 styles, Gemini, Atlas Cloud, or MuAPI AI), corporate identity program (50 deliverables, CIP mockups), HTML presentations (Chart.js), banner design (22 styles, social/ads/web/print), icon design (15 styles, SVG…
Languages
Trust signals
From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.
