How do I install the Freesearch Search skill?

Run npx skillmds@latest add kennethkhoocy/freesearch-search in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Freesearch Search safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Freesearch Search?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Freesearch Search free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Freesearch Search?

kennethkhoocy (@kennethkhoocy) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Freesearch Search by kennethkhoocy — Agent Skill

Name: Freesearch Search
Author: kennethkhoocy

Freesearch Search

Run the lit-review orchestrator free index search channel against keyless OpenAlex, Crossref, and Semantic Scholar keyword-search endpoints. Use when the user invokes the free search channel, asks for keyless scholarly index search, or needs Stage 4e fallback discovery without SearchAPI, Gemini, or Undermind credentials.

kennethkhoocy 06abc4f 2 files · 14.2 KB Updated Sep 17, 2026

File contents

Free Index Search channel (Stage 4e) — keyless OpenAlex / Crossref / Semantic Scholar

A zero-key discovery channel: it runs the Stage-0 query list against the free, keyless keyword-search endpoints of OpenAlex, Crossref, and Semantic Scholar and normalizes the real records into the pipeline schema. Together with the agent web-search channel (Stage 4d) it is the "only Claude Code" search fallback — no search account or API key required.

Unlike web search, this is a plain subprocess (it calls keyless HTTP APIs), so it runs the same way in any context.

When to use

The user has no SEARCHAPI_API_KEY / GEMINI_API_KEY / Undermind login — this is the keyless search backbone, OR
You want extra grounded, clean-metadata coverage from the scholarly indexes alongside the other channels.

Run it in parallel with whatever else is available; its output merges at dedup.

Recipe

Use the same Stage-0 query list Google Scholar uses:

python freesearch-search/scripts/freesearch_search.py \
    --queries-file OUT/scholar_queries.json -o OUT/stage4e_freesearch.json

Or a single query with --query "...". Options: --per-query N (default 20) and --sources openalex,crossref,semanticscholar (any subset). It writes stage4e_freesearch.json (+ .ris) with source set per index, deduped across indexes by DOI then title. The stage[0-9]*.json dedup glob picks it up automatically.

Keys

None required. OpenAlex and Crossref use the polite pool via a mailto (set LITREVIEW_CONTACT_EMAIL to use your own contact). If present, OPENALEX_API_KEY and SEMANTIC_SCHOLAR_API_KEY are used only to raise rate limits. It makes no LLM API calls.

Anti-hallucination

Every hit is a real index record, so there is nothing to fabricate — relevance, not existence, is the only variable, and screening handles relevance. Stage 5b verification still confirms each paper downstream. The keyless Semantic Scholar tier is rate-limited, so it is queried with a short delay and skipped gracefully on a 429.

---name: freesearch-searchdescription: Run the lit-review orchestrator free index search channel against keyless OpenAlex, Crossref, and Semantic Scholar keyword-search endpoints. Use when the user invokes the free search channel, asks for keyless scholarly index search, or needs Stage 4e fallback discovery without SearchAPI, Gemini, or Undermind credentials.---# Free Index Search channel (Stage 4e) — keyless OpenAlex / Crossref / Semantic ScholarA zero-key discovery channel: it runs the Stage-0 query list against the **free,keyless keyword-search endpoints** of OpenAlex, Crossref, and Semantic Scholar andnormalizes the real records into the pipeline schema. Together with the agentweb-search channel (Stage 4d) it is the "only Claude Code" search fallback — nosearch account or API key required.Unlike web search, this is a plain subprocess (it calls keyless HTTP APIs), so itruns the same way in any context.## When to use- The user has no `SEARCHAPI_API_KEY` / `GEMINI_API_KEY` / Undermind login — this is the keyless search backbone, OR- You want extra grounded, clean-metadata coverage from the scholarly indexes alongside the other channels.Run it in parallel with whatever else is available; its output merges at dedup.## RecipeUse the same Stage-0 query list Google Scholar uses:```bashpython freesearch-search/scripts/freesearch_search.py \ --queries-file OUT/scholar_queries.json -o OUT/stage4e_freesearch.json```Or a single query with `--query "..."`. Options: `--per-query N` (default 20) and`--sources openalex,crossref,semanticscholar` (any subset). It writes`stage4e_freesearch.json` (+ `.ris`) with `source` set per index, deduped acrossindexes by DOI then title. The `stage[0-9]*.json` dedup glob picks it upautomatically.## KeysNone required. OpenAlex and Crossref use the polite pool via a mailto (set`LITREVIEW_CONTACT_EMAIL` to use your own contact). If present, `OPENALEX_API_KEY`and `SEMANTIC_SCHOLAR_API_KEY` are used only to raise rate limits. It makes no LLMAPI calls.## Anti-hallucinationEvery hit is a real index record, so there is nothing to fabricate — relevance, notexistence, is the only variable, and screening handles relevance. Stage 5bverification still confirms each paper downstream. The keyless Semantic Scholar tieris rate-limited, so it is queried with a short delay and skipped gracefully on a 429.

Frequently asked questions