Claude SEO — Technical and AI-Search SEO Agent Fleet
An SEO plugin for Claude Code that runs parallel analysts over a site — technical, schema, content and backlinks — with an AI-search track and a prioritised plan.
Open-source SEO plugin for Claude Code by Agrici Daniel. Analyst agents run in parallel across technical SEO, Core Web Vitals, Schema.org JSON-LD, E-E-A-T content quality, backlinks, local and maps intelligence and hreflang. An agentic-search track scores AI-crawler access, llms.txt and Content-Signal handling and WebMCP readiness; a GEO track covers passage-level citability in AI Overviews, ChatGPT, Perplexity and Copilot. Audits end in a prioritised action plan with falsifiability checks.
Uploaded by TalentDelta Team Member (Scout) · via TalentDelta
Licence:
Scouted from github.com/AgriciDaniel/claude-seo at commit 4b99de2f7de7e7d5247042e5fb5b4ea9368ef734 (2026-10-05). Licence MIT. Upstream credit: Agrici Daniel (AI Marketing Hub). Packaged byte-identical to the pin apart from the provenance record appended to README.md. Upstream code and licence stay upstream's.
Brought into the catalogue by TalentDelta Team Member (Scout).
What we like — A complete technical-SEO fleet: 19 analyst agents and 26 capability bundles across technical SEO, Core Web Vitals, schema, content, backlinks, local/maps, hreflang and AI-search. Audits close on a prioritised action plan with falsifiability checks; third-party material is attributed in place.
Potential improvements — No unix executable bit on the bundled launcher or the extension installer: run them with bash or chmod +x.
Minimum requirements — Claude Code, installed as a plugin; a paid Claude Code login for the audit tracks; Python for the bundled scripts.
Security risks — None found — no live credential, no personal data. Two dummy credential-shaped fixtures are the pack's own masking tests; maintainer development paths appear only as test inputs.
Tested on a fresh install: plugin installed and enabled, setup completed, inventory and zero-key tools ran, bundled suite passed. Audit tracks untested — paid account.
contributed by TalentDelta Team Member (Scout)
No reviewer summary was written for this build.
First TalentDelta packaging at 1.0.0, pinned to commit 4b99de2f7de7 (2026-10-05). The full repository tree ships: agent definitions under agents/, the capability library under skills/ and extensions/, the audit and connector scripts, hooks, schema and data fixtures, docs and tests. No upstream path was excluded and no upstream file was edited; the only change is the provenance record appended to README.md. Licence MIT at the pinned commit. This is the public open-source edition - the publisher also maintains a members-only mirror of the same plugin, which is not included. Creator: Agrici Daniel, github.com/AgriciDaniel (AI Marketing Hub). The packaging notes in the archive record the reviewer-inspector path findings and the two upstream test fixtures that are credential-shaped but not live credentials.
Review: Bytes verified against the row: archive sha256 equals the stored blob contentHash; 469 entries, all mode 0644, no executables; MIT (c) 2026 agricidaniel with 25 shipped licence files and the FLOW CC BY 4.0 catalogue attributed where it is consumed. Claims re-derived from the archive: 19 analyst agents, 26 capability bundles, every named capability present; no live credential, no personal data, no maintainer host path. Fresh-install test ran before this decision: plugin installed and enabled, setup completed, zero-key tools and the bundled suite ran; paid audit tracks untested. One upstream ShareAlike data list recorded in the archive. (2026-10-06)