Add Semantic Scholar no-auth chat tools app

[liangtovi-debug]

/claim #3120

Summary

• add a standalone plugins/omi-semantic-scholar-app integration with no OAuth/API key requirement
• expose 3 chat tools:
  • search_semantic_scholar_papers
  • get_semantic_scholar_paper
  • get_semantic_scholar_author_papers
• follow Omi chat-tool patterns:
  • JSON body input models via Pydantic
  • ChatToolResponse using result/error
  • manifest at /.well-known/omi-tools.json
• include deploy/runtime files (requirements.txt, Procfile, railway.toml) and README

Validation

• python3 -m py_compile plugins/omi-semantic-scholar-app/main.py plugins/omi-semantic-scholar-app/models.py

Notes

• Uses public Semantic Scholar Graph API endpoints only (no credentials).
• ID lookups URL-encode user-provided identifier path segments before requests.

[greptile-apps]

Greptile Summary

Adds a new standalone Omi plugin (plugins/omi-semantic-scholar-app) that exposes three no-auth chat tools — paper search, paper detail lookup, and author-papers lookup — backed by the public Semantic Scholar Graph API. The plugin follows FastAPI + Pydantic patterns used by other plugins in the repo.

• normalize_identifier strips the doi: prefix before building the API path, so user-supplied DOIs like doi:10.1234/5678 arrive at Semantic Scholar as 10.1234%2F5678 (no DOI: prefix), which the API cannot resolve and returns a 404; additionally quote(..., safe=\"\") encodes the colon in any preserved prefix (e.g. arXiv:), compounding the issue.
• Manifest uses plain-string parameter values instead of the JSON Schema shape (properties/type/description/required) that other plugins in this repo use, which may prevent Omi from correctly passing typed arguments to these tools.
• ChatToolResponse allows both result and error to be None simultaneously with no validator enforcing that at least one is populated.

Confidence Score: 3/5

The paper-by-DOI lookup path is broken out of the box — every DOI-based request will return a 404 from Semantic Scholar — so one of the three advertised tools is non-functional as written.

The normalize_identifier function strips the doi: prefix that Semantic Scholar requires to distinguish a DOI from a native paper ID, and quote(..., safe="") would additionally encode any colon in other identifier schemes (e.g., arXiv:). Together these mean the get_semantic_scholar_paper tool silently fails for any non-native-ID input. The other two tools (search and author lookup) are functionally correct, but the manifest's non-standard parameter format may prevent the Omi platform from properly wiring tool arguments.

main.py — specifically normalize_identifier (lines 38–42) and the quote call in get_paper (line 124), and the manifest parameter schema (lines 57–81).

Important Files Changed

Filename	Overview
plugins/omi-semantic-scholar-app/main.py	Core app logic with two bugs: normalize_identifier strips the DOI prefix needed by Semantic Scholar, and quote(..., safe="") encodes colons, breaking all DOI-based paper lookups. Manifest also uses non-standard parameter format.
plugins/omi-semantic-scholar-app/models.py	Pydantic models are well-structured; ChatToolResponse allows both result and error to be None simultaneously with no validation guard.
plugins/omi-semantic-scholar-app/requirements.txt	Pinned dependency versions for fastapi, uvicorn, httpx, and pydantic; no issues found.
plugins/omi-semantic-scholar-app/Procfile	Standard Heroku/Railway Procfile for uvicorn startup; no issues.
plugins/omi-semantic-scholar-app/railway.toml	Railway deployment config with health check and restart policy; no issues.
plugins/omi-semantic-scholar-app/README.md	Minimal README covering local run and manifest URL; no issues.

Sequence Diagram

sequenceDiagram
    participant Omi as Omi Platform
    participant App as omi-semantic-scholar-app (FastAPI)
    participant SS as Semantic Scholar Graph API

    Omi->>App: GET /.well-known/omi-tools.json
    App-->>Omi: Tool manifest (3 tools)

    Omi->>App: POST /tools/search_semantic_scholar_papers
    App->>SS: "GET /paper/search?query=...&limit=...&fields=..."
    SS-->>App: "{data: [...papers]}"
    App-->>Omi: "ChatToolResponse {result: formatted list}"

    Omi->>App: POST /tools/get_semantic_scholar_paper
    Note over App: normalize_identifier() strips doi: prefix
    Note over App: quote(..., safe="") encodes colon — DOI lookup breaks
    App->>SS: "GET /paper/{encoded_id}?fields=..."
    SS-->>App: paper detail or 404
    App-->>Omi: "ChatToolResponse {result or error}"

    Omi->>App: POST /tools/get_semantic_scholar_author_papers
    App->>SS: "GET /author/{encoded_id}?fields=name,papers.*"
    SS-->>App: "{name, papers: [...]}"
    App-->>Omi: "ChatToolResponse {result: sorted paper list}"

Loading

_{Reviews (1): Last reviewed commit: "Add Semantic Scholar no-auth chat tools ..." | Re-trigger Greptile}

[greptile-apps]

DOI prefix stripped and colon encoding — DOI lookups always 404

normalize_identifier removes the doi: prefix before the value is passed to the Semantic Scholar API. Semantic Scholar's paper lookup endpoint requires the DOI: prefix to recognize the identifier as a DOI (e.g., /paper/DOI:10.1234%2F5678); without it the raw value 10.1234%2F5678 is treated as an unknown paper ID and the API returns a 404.

Compounding this, quote(..., safe="") on line 124 encodes : to %3A, so even if the prefix were preserved, DOI%3A10.1234%2F5678 may not be matched by the server. The fix is to normalise the prefix to uppercase DOI: (instead of stripping it) and pass safe=":" to quote so that scheme separators are left intact.

[greptile-apps]

Non-standard manifest parameter schema — Omi may not parse tool parameters correctly

Other plugins in this repo (e.g., omi-zomato-app) use a JSON Schema-shaped parameters object with properties, type, description, and required keys. The current manifest uses plain strings as values (e.g., "string", "integer 1-10 (default 5)"). If Omi's tool-calling layer expects JSON Schema, the LLM will not receive accurate type or constraint information for these tools.

Suggested change

	{
	"name": "search_semantic_scholar_papers",
	"description": "Search Semantic Scholar papers by keyword.",
	"endpoint": "/tools/search_semantic_scholar_papers",
	"method": "POST",
	"parameters": {
	"query": "string",
	"max_results": "integer 1-10 (default 5)",
	"min_year": "optional integer",
	},
	},
	{
	"name": "get_semantic_scholar_paper",
	"description": "Get details for a paper by Semantic Scholar ID or DOI.",
	"endpoint": "/tools/get_semantic_scholar_paper",
	"method": "POST",
	"parameters": {"paper_id_or_doi": "string"},
	},
	{
	"name": "get_semantic_scholar_author_papers",
	"description": "Get recent papers by Semantic Scholar author ID.",
	"endpoint": "/tools/get_semantic_scholar_author_papers",
	"method": "POST",
	"parameters": {"author_id": "string", "max_results": "integer 1-10"},
	},
	{
	"name": "search_semantic_scholar_papers",
	"description": "Search Semantic Scholar papers by keyword.",
	"endpoint": "/tools/search_semantic_scholar_papers",
	"method": "POST",
	"parameters": {
	"properties": {
	"query": {"type": "string", "description": "Keyword search query"},
	"max_results": {"type": "integer", "description": "Number of results to return (1-10, default 5)"},
	"min_year": {"type": "integer", "description": "Optional earliest publication year filter"},
	},
	"required": ["query"],
	},
	},
	{
	"name": "get_semantic_scholar_paper",
	"description": "Get details for a paper by Semantic Scholar ID or DOI.",
	"endpoint": "/tools/get_semantic_scholar_paper",
	"method": "POST",
	"parameters": {
	"properties": {
	"paper_id_or_doi": {"type": "string", "description": "Semantic Scholar paper ID or DOI (e.g. DOI:10.xxx/xxx)"},
	},
	"required": ["paper_id_or_doi"],
	},
	},
	{
	"name": "get_semantic_scholar_author_papers",
	"description": "Get recent papers by Semantic Scholar author ID.",
	"endpoint": "/tools/get_semantic_scholar_author_papers",
	"method": "POST",
	"parameters": {
	"properties": {
	"author_id": {"type": "string", "description": "Semantic Scholar author ID"},
	"max_results": {"type": "integer", "description": "Number of papers to return (1-10, default 5)"},
	},
	"required": ["author_id"],
	},
	},

[greptile-apps]

ChatToolResponse allows both fields to be None simultaneously

Both result and error default to None, so it is possible to construct — or accidentally return — ChatToolResponse() with {"result": null, "error": null}. Adding a model validator (e.g., @model_validator(mode="after")) that requires exactly one of the two fields to be set would make the contract explicit and prevent silent empty responses from reaching the Omi platform.

[liangtovi-debug]

Addressed Greptile findings in c2099ae:

1. Fixed DOI handling in get_semantic_scholar_paper:

• Preserve DOI namespace as DOI:... in normalization.
• Keep colon safe while URL-encoding path segment (quote(..., safe=":")) so DOI prefix survives and slash is encoded.

2. Updated manifest parameter definitions to JSON Schema object format (type/properties/required) for all three tools, matching existing Omi plugin conventions.

3. Added ChatToolResponse model validation to ensure at least one of result or error is present.

Validation:

• python3 -m py_compile plugins/omi-semantic-scholar-app/main.py plugins/omi-semantic-scholar-app/models.py ✅
• Live Semantic Scholar DOI smoke call attempted, currently rate-limited by upstream (HTTP 429), so no deterministic live API assertion in CI here.

[kodjima33]

thanks for adding the Semantic Scholar app