Add BotSee AI visibility tool

[sidekickjanet]

Summary

• Add BotSee to the Marketing & SEO category as an agent-native AI-visibility measurement workflow.
• Document the hosted API, Claude Code plugin, and direct Python CLI using first-party documentation and the maintained plugin repository.
• Keep the listing factual: structured competitor, keyword, source, and raw-response records are returned for review; no performance or coverage claims are made.

Inclusion rationale

BotSee makes agents first-class participants in a documented workflow: an agent configures a site and buyer-question benchmark, triggers analysis, and retrieves structured records programmatically. This is substantive measurement infrastructure for agent-led marketing and AI-search research, not generic API compatibility.

Validation

• npm run validate:content -- --require-submitters
• npm test (84 passing)

[foo-bender]

Outcome: changes-requested

Policy assessment
@sidekickjanet, BotSee's documented agent benchmarks and structured keyword/source/raw-response retrieval establish material agent value beyond generic compatibility; agent-friendly / marketing-seo is reasonable. The complete frontmatter/evidence/contradiction pass for tools/botsee.md found these blockers:

• Reconcile pricing across current first-party surfaces before asserting pay-per-report without a subscription. https://botsee.io/ and https://botsee.io/docs/cli describe $129/$159/$199 one-time reports, but https://botsee.io/pricing shows $129/week, $159/week and $199/week. Identify the specific edition/channel each applies to, obtain current first-party clarification, and qualify or remove the conflicting no-subscription claim in classification rationale, limitations and body. Do not simply choose the cheaper supporting page.
• githubUrl points at a real client/plugin implementation, not the hosted analysis backend described by this record. The repository explicitly excludes that private backend. Remove that field for the hosted product, or consistently scope/name the record as the CLI/client and separate its source/license from the hosted service. The GitHub link may remain in evidence with that explicit distinction.
• Add material constraints: Claude Code plugin vs direct Python CLI for Codex/other coding agents; paid reports and annual API-key access are distinct modes (FAQ: $4,999/year or $999 with a report); report retrieval tokens operate as bearer capabilities (HTTPS/SSE transport is not end-to-end encryption; stored data is disclosed as not encrypted); third-party LLM handling; and retention differences (archived reports/history/raw-response files vs 30-day report context).
• Reconcile current version/provider and refresh claims. Main docs/GitHub contain older provider/default-count and plugin-version examples and CLI refresh cadence, while current installation/changelog/pricing pages show a newer plugin and weekly/monthly refresh by package. Delimit CLI report vs hosted subscription behavior; do not present one channel's cadence or historical provider set as universal.

Evidence checked
Both exact evidence URLs support their bounded agent-interface/client claims:
https://botsee.io/docs
https://github.com/RivalSee/botsee-skill
The independent contradiction pass additionally checked the official homepage, pricing, CLI/help/installation/changelog/API/session/tool docs, FAQ, security, privacy, repository README/implementation/manifest/license, and release/package metadata. The client is MIT-licensed; no hosted-backend open-source claim is supported. Runtime model accuracy, payments and purchased reports were not tested.

Checks run
npm run validate:content: passed on exact head merged with main 273dfa6 (54 tools, 18 categories).
npm test: passed, 90/90. Current GitHub enrichment check passed. No SEO evaluation or metadata commit was applied; resolve these substantive conflicts first.