| name | string-ppi-search |
| description | Query STRING for protein-protein interactions, interaction confidence/evidence, interaction partners, and functional enrichment. Use for PPI networks and enrichment of a protein set. Public API, no credential needed. |
| license | Apache-2.0 |
String Ppi Search (Codex-native)
Codex-native adaptation of Google DeepMind's string-database skill
(science-skills, Apache-2.0).
Uses the public API directly through Codex's own tools.
Gates (ask before proceeding)
- Terms notice: on first use in a workspace, tell the user to review the
source's terms of use and record the notice in provenance.
- Network / install: read-only public access; be polite about rate limits.
Write large responses to a file and parse them; if you write helper code, run
it via
uv run, never bare python3. Ask before installs or large downloads.
Endpoints
- STRING API
https://string-db.org/api (e.g. json/network, json/interaction_partners, json/enrichment)
Workflow
Resolve a protein first with science_search_string; use STRING directly for
species-constrained networks, partners, enrichment, and multi-protein queries.
- Map identifiers first (
/json/get_string_ids), then request the network/partners with a species id and a required_score threshold.
- For enrichment, POST the identifier set to
/json/enrichment.
- Provenance & review: record queries, endpoints, and cited identifiers with
$science-provenance; check claims with $science-review before presenting.
Boundaries
- Report the species and confidence threshold; STRING combines predicted and experimental evidence.
- Interactions include predictions; distinguish evidence channels. Cite only ids actually returned.
