edit.pink
Pink Editor docs

Tool reference

All 39 tools Pink Editor's MCP server offers, grouped by what they do.

API v1

Pink Editor's MCP server offers 39 tools. Your AI client lists them when it connects (tools/list); the pages in this section show each tool's description and input schema as the client receives them.

Costs are on Credits and pricing. Tools not listed there are free.

Sites, session and memory

ToolWhat it does
list_sitesList the user's Pink Editor sites so the operator can pick one.
select_siteBind the chosen site to this MCP session AND get the foundation back in the same response.
add_siteCreate a brand-new Pink Editor site and immediately switch this MCP session to it (same effect as select_site).
get_siteReturn the currently-selected site_id (or null if none chosen).
get_foundationRe-fetch the foundation (brand essence only) for the bound site: site, audiences, goals, ctas, writable_cols, skills_registry, memory_registry, today, current_selection, access.
get_selectionReturns the page(s) and/or topic the user has currently selected/clicked in the Pink Editor UI, with details (keywords, pains, questions, notes) pre-joined so you can act without …
get_skillLoad a named playbook for a workflow.
set_memoryAppend a bullet to the site's memory document, or overwrite the whole document.
undoUndo/redo recent mutations.

Content plan

ToolWhat it does
graphRead-only typed query over the site's content graph.
recallSemantic search over the site's memory.
db_selectRead rows from any cm_* table, site-scoped automatically (cm_keywords and cm_research are shared pools, not per-site).
db_insertInsert rows into a writable cm_* table.
db_updateUpdate rows matching WHERE with patch.
db_deleteDelete PLANNING rows matching WHERE.
set_pageUpdate a cm_pages row.
set_linksAuthor internal-link creates/deletes (single or bulk).
set_keyword_rolesAtomic primary/secondary/candidate role swap on cm_page_keywords.
merge_topicsMerge source topics INTO target topic: re-parents pages, moves keyword attachments, soft-deletes sources.

Research, keywords and SERP

ToolWhat it does
researchFour modes: keywords (seed is 3–5 short seeds, each 1–3 words, comma-separated; returns volume/CPC in USD (cpc_usd)/intent), web (web search + scrape + synthesis of …
get_keywordsPull fresh keyword metrics (volume/CPC/competition) for seeds[], topic_id, or keyword_id.
serp_landscapeThe keyword + competitor intelligence tool.
serp_backlinksBacklink profiles — ONE tool, the scope arg picks the shape (like serp_landscape's depth).
serp_keyword_refreshForce a fresh single-keyword SERP pull (capped at 20/day/site) — the current top-20 organic results plus any untracked competitors that show up there.
add_competitorTrack a competitor domain so it appears in the keyword-landscape and backlinks views with its share-of-voice and backlink profile populated.

Search Console

ToolWhat it does
gsc_query_statsGet Search Console performance for a specific page: clicks, impressions, avg position, top queries over a recent window.
gsc_striking_distancePage+query pairs in positions 8–20 with real impressions — small edits typically push them into the top 5.
gsc_content_decayPages whose clicks/day dropped >40% comparing last 28d vs prior 62d.
gsc_cannibalizationQueries where 2+ of our URLs compete in the 15–25 band and none wins.
gsc_hidden_winnersPages where Google has hidden most of the query data (anonymized impressions ≥40%).
gsc_forecast90-day click forecast per page with confidence intervals.

Writing and publishing

ToolWhat it does
get_article_briefReturn everything the LLM needs to understand one cm_pages row before publishing: foundation (brand/audiences/goals/CTAs — the brand voice is not included, it is applied …
add_articlePublish a page (Article, FAQ, Landing, Pillar, or Utility).
get_pipeline_statusCheck the current article-pipeline status for a page.
get_articleRead a published article before changing it.
set_articleReplace a published article's prose, its status, or both.
get_answerRead a Q&A page before changing it.
set_answerReplace a Q&A answer, its status, or both.
add_answerGenerate and publish the answer for ONE researched question, without running the whole add_article pipeline.

On this page