MimikFlow

MimikFlow's optional Contenu option (bought per LinkedIn account) prepares the user's own LinkedIn posts every week, places them in a calendar and publishes them from the connected LinkedIn account. Reads ONE post in full and SHOWS its picture: the hooks proposed (`hooks`, pick one with write_content_post), the retained hook, the body, `final_text` (exactly what goes to LinkedIn: hook, blank line, body), the call to action, the slot (UTC and account time zone), the review (`editor.score` out of 10 and its findings), `facts_flag` (claims the machine could not back with the user's facts: check them with the user before approving), the publication state (`last_error`, `share_url`, metrics once published) and the visual: its kind, the layout and text of a drawn visual, the photo used, the photos and web images proposed (`web_index` for set_content_post_visual). With `include_image` (default true) the current visual follows as an image; with `include_candidates` the proposed photos and web images follow too (up to 8 images). Describe what you see when the user asks, and judge the visual against the text. Codes and ids are for tool calls only: to the user say 'faire connaître' / 'get known' (tofu), 'faire comprendre' / 'get understood' (mofu), 'faire acheter' / 'get bought' (bofu), and the post types in words.

MimikFlow's optional Contenu option (bought per LinkedIn account) prepares the user's own LinkedIn posts every week, places them in a calendar and publishes them from the connected LinkedIn account. Reads what the posts of one account brought and what the watch found. `funnel_4w`: the real split of the posts published or approved over four weeks, by funnel stage. `published`: posts sent, with reactions, comments, reposts, impressions and `leads_from_reactions` (people who reacted and entered prospecting); failed or unconfirmed sends appear with their status. `what_works`: the lessons learned from the account's posts, each with its `line_id` and whether the user kept it (act on one with update_content_setup). `old_posts`: the account's past LinkedIn posts, best relative engagement first, paginated (`search` keeps those containing a word). `brief`: the latest weekly brief (trends and why they matter to the user, dated news, the prospects' voice, anonymised, and the funnel debt: positive = that stage is behind). `competitor_insights`: what the competitors' recent posts teach (what works for them, what they do, what to take), each line with the `evidence_ids` of inspirations proving it, null before the first analysis. `inspirations`: competitors' posts, trending posts and news of the last `days` days, with their source. `journal`: what the machine and the user did, most recent first. `posts_that_work`: the day's best recent LinkedIn posts in the account's language, chosen every day among what MimikFlow reads for all its users, by post type, each with what makes its form work; when you write a post for the user, follow their length, airing, opening and ending in the user's own voice, never their subject, sentences or numbers. Impressions are views, not prospects; a prospect from a reaction is not a sale.

MimikFlow's optional Contenu option (bought per LinkedIn account) prepares the user's own LinkedIn posts every week, places them in a calendar and publishes them from the connected LinkedIn account. Reads how one account is set up. `settings`: what update_content_setup writes: mode (validation = the user approves every post, the default; auto = clean posts are approved by the machine), posts_per_week, proposals_per_week, days_csv, the hours the machine picks (hour_start, hour_end), plan_weekday (0 = Monday) and plan_hour of the weekly preparation, language, identity_text, convictions_text, proof_points (the ONLY facts a post may state), signatures, territories, funnel_mix, intents_enabled, voice (register, banned_words, tics, standing_instructions every post respects), watched_handles, keywords, news_topics, image_policy and `strategy_doc` (positioning, mechanism, positions = the author's stances, audience, pillars, editorial line, rhythm, watch, goals, lead_magnets = the ONLY resources a post may offer, ctas = the calls to action the user wants posts to end with, excluded_campaigns = the campaigns the posts must not talk about). `from_campaigns`: READ from the account's active campaigns the posts talk about (offer, target, objections, proof, goal), changed in the campaign, never here; `campaign_choices` lists every active campaign with `included`. `visual_style`: the look of the visuals drawn from a post's words (colors, font, the author's name and line, logo, web image preference), the fonts and the two ready palettes offered (light by default, night), and with `include_previews` the four layouts drawn in that style (the sentence, the figure, the photo, the quote). `photos`: the photo library for posts (id, what it shows, when to use it; a photo without a description is never proposed). With `include_photo_images` the photos follow as images (8 images at most in all): you may suggest a photo description from what you see, saved only once the user confirms it.

Full detail of one prospect: LinkedIn profile info, status, and the complete conversation thread across every channel MimikFlow uses with them (LinkedIn and, when the number is connected, WhatsApp), plus any draft waiting for approval. Each message says who wrote it: 'prospect' for their replies, 'us' for what left the user's account, with `written_by` telling apart the AI and the user's own hand-typed messages. BEFORE reading, this tool REFRESHES the live LinkedIn thread, so what it returns includes messages the user typed by hand on LinkedIn a minute ago. ALWAYS call it immediately before approving a draft or answering a prospect -- it is the only way to know nobody already answered. `awaiting_reply_from` says who owes the next message: 'us' means the prospect is waiting, 'prospect' means WE spoke last and writing again would be a second message in a row. Long threads are cut to the 100 most recent messages: when `truncated` is true the thread STARTS EARLIER than what you see (`total_messages` = the real count), so never conclude anything about how the conversation began. `status_set_by` says who put the prospect in their current status: 'human' means somebody moved them by hand in the app, 'ai' means the machine decided it, 'unknown' means nothing was recorded. Use it before calling a status surprising: a won prospect with no conversation was almost always moved by hand, and saying the machine did it is wrong. Pass refresh=false only to re-read a thread you refreshed seconds ago.

Journal of the user's CRM integration (HubSpot, Pipedrive or Folk connected to MimikFlow, their automated LinkedIn prospecting tool): each line is one plain-language event, newest first. Kinds: `pushed` / `push_updated` (a hot prospect was created or updated in the CRM: the user's CRM receives a prospect when they reply, book a meeting or reach the goal), `push_failed` (could not be sent after a day of retries), `exclude_synced` (the never-contact list was refreshed from the CRM's customers and open deals), `exclude_stopped_leads` (prospects in progress the CRM now excludes, they receive nothing more), `listen_applied` (a deal won or lost in the CRM stopped the prospecting for that person), `auth_lost` (the CRM refused access: the user must reconnect it in Settings > CRM). Read-only: this connector never writes into the CRM and never changes its settings; send the user to the app for that. Use `since_days` for questions like 'what went to HubSpot this week'. Paginated: check `has_more` before saying you saw everything.

Create the secure LinkedIn connection or reconnection link used by MimikFlow's existing onboarding flow. Call this only when get_onboarding_status says connect_linkedin or reconnect_linkedin. Give the returned link to the user, then call get_onboarding_status again after they finish; never claim the account is connected merely because the link was created. This action sends no invitation or message and does not activate prospecting.

Read the user's MimikFlow setup and return exactly what to do next. Use this first when the user wants to start prospecting: it reports whether their offer and ideal customer are saved, whether LinkedIn is connected, the missing fields, and one next_action. It never changes settings or starts outreach. If setup is complete, continue with search_prospects for a small preview only; never add or contact anyone without a separate confirmation.

Read a campaign's fine-grained machine settings: reminder cadence (timing_* waits, timing_max_follow_ups, the selected offer-pitch follow-up, per-phase relance styles), first-message tone (cold_dm_formality tu/vous, style, length, emoji, CTA mode), setter tone (setter_style, setter_max_phrases), sending modes (cold_dm_mode / setter_mode: 'auto' sends by itself, 'semi_manual' queues every draft for the user's approval), sending days/hours per pipeline, daily/weekly volume caps, market/language settings, and the meeting follow-through (Suivi RDV) keys: nurture_pre_meeting / nurture_post_meeting (opt-in value messages before/after a booked meeting), meeting_reminder + meeting_noshow_recovery + meeting_cancel_recovery (live defaults, 'off' disables), and setter_post_booking (on by default, 'off' disables: the AI answers booked prospects' questions, stays silent on a thanks or an ok, and a prospect who cancels or postpones is re-booked or handed to the human). The meeting keys are the most often left untouched and the most valuable to set up: a reminder before the call and a message adapted to whether the prospect showed up are what turn booked calls into calls that happen, and `meeting_touches` can carry a media (a demo, a case study -- see list_media). Every empty value simply means MimikFlow's sensible default applies (these are tuning knobs, not required configuration). The keys are internal vocabulary: to the user, name each setting by what it does in their language ('le rythme des relances', 'le tutoiement du premier message'), never by its raw key. Read before updating with update_pipeline_settings.

The user's LinkedIn prospecting KPIs and funnel for a period: connection invites sent/accepted, first messages (cold DMs) sent, prospect replies, conversion goals reached (goal_reached = a won lead, usually a booked meeting), with per-step drop rates. Use it for recaps ('how is my prospecting going?') and to diagnose where the funnel leaks. A frequent leak: invites accepted but few replies, which usually means the first message's approach (formality, style, hook) does not fit the audience -- read real threads and the campaign profile, then tune the cold_dm_* settings. window is '7d', '30d' or 'lifetime'. Report the numbers to the user in plain words in their language (invitations envoyées / acceptées, premiers messages, réponses, objectifs atteints), never as raw field names.

Read one campaign's WhatsApp settings: the wording of the first message, the reminders and the AI that answers on this channel, its sending days/hours, plus the plan of messages around a booked meeting (see update_meeting_moments). What the campaign SAYS -- the offer, the target, the goal, the voice, the proof, the media library -- is shared with LinkedIn and is read with get_campaign_profile: only the channel's own wording lives here. An empty value means MimikFlow's default applies, which is a working configuration, never a gap. Name each setting to the user by what it does in their language, never by its key.

Raw material to coach the user on their sales conversations: full threads of recently WON leads (conversion goal reached) and LOST leads (dropped after the prospect had replied). YOU do the analysis: recurring objections, drop-off points, what worked in wins. Then translate EVERY finding into a concrete settings change (update_targeting for who gets contacted; update_campaign_profile for my_objections, the custom instructions, style or proof; update_pipeline_settings for the first-message tone when the opener does not fit the profiles contacted) and offer to apply it. Frame findings as settings to improve, never as defects of the product. outcome is 'won', 'lost' or 'both'.

The user's connected LinkedIn accounts. Most users have ONE; an AGENCE user runs several (e.g. one per client) from this single connector. Returns each account's id, name (label), LinkedIn plan, status (active/disconnected), how many campaigns run on it, and its lead-dedup pool (accounts sharing a non-empty pool never prospect the same people). Use this to know which accounts exist, then list_campaigns to see the campaigns of each, and target any campaign by its campaign_id -- one connector pilots every account.

The user's prospecting campaigns. A campaign bundles one ideal customer profile (ICP) + one offer + one tone + its own leads and conversations; most users have one, agencies run several in parallel. Returns id, name, status (active/archived), lead count, and the LinkedIn account it runs on (linkedin_account_id + account_name -- an Agence user targets a client by picking that client's campaign). Optional `account_id` filters to one account's campaigns. When a tool gets no campaign_id it uses the primary (oldest active) one.

MimikFlow's optional Contenu option (bought per LinkedIn account) prepares the user's own LinkedIn posts every week and publishes them from the connected account. Its idea notebook holds the post ideas the user wants written: the weekly preparation reads it FIRST. In the app the notebook lives in the Pilote chat of the Contenu page. Returns, for one LinkedIn account, `notebook`: the ideas waiting to be written, newest first and paginated (`total`, `has_more` and `next_offset` describe the notebook: page before counting it). `recently_used` (already written into a post) and `recently_dismissed` (set aside) show the latest ones with their own `total`, for context. `weekly_preparation` gives `next_run_at` (account time zone) and `posts_need_user_approval` (false = posts go out without a manual review), or a `note` saying why it will not run. Each idea: id (for manage_content_idea), title, thesis (the one-sentence point of the post), angle, stage, post_type, source ('user' = the user's own, anything else = suggested by MimikFlow) and created date. `stage` (funnel stage): tofu = 'faire connaître' / 'get known', mofu = 'faire comprendre' / 'get understood', bofu = 'faire acheter' / 'get bought'. `post_type`: casser_croyance = 'casser une croyance' / 'break a belief', eduquer = 'éduquer' / 'educate', prouver = 'prouver' / 'prove', expliquer_secteur = 'expliquer le secteur' / 'explain the sector', actualite_commentee = 'actualité commentée' / 'news with an opinion', coulisses_histoire = 'coulisses et histoire' / 'behind the scenes', offre_directe = 'offre directe' / 'direct offer'. Both are optional: left empty, the weekly preparation chooses. Codes and ids are for tool calls only: to the user say the quoted words, 'ton carnet d'idées' / 'your idea notebook' and 'la préparation de la semaine' / 'the weekly preparation'.

MimikFlow's optional Contenu option (bought per LinkedIn account) prepares the user's own LinkedIn posts every week, places them in a calendar and publishes them from the connected LinkedIn account. Lists the posts of one account as cards: id, status (draft = waiting for the user's approval, approved = will publish at its slot, publishing, published, unknown_outcome = sent but not confirmed yet, failed, rejected, expired), funnel stage and post type, hook, a short excerpt, the slot in UTC and in the account's time zone (`slot_local`), flags (`review_flag` = to read again, `facts_flag` = claims to check before publishing, `awaiting_validation` = missed two slots and waits for the user), the kind of visual. Filters: `from` / `to` (ISO dates, `to` included) keep the posts whose slot falls in that range, earliest first; `reserve=true` keeps the proposals without a slot; `status` narrows. Without a filter: every post written, newest first. Paginated (`total`, `has_more`, `next_offset`). Read one post in full, with its image, with get_content_post. Codes and ids are for tool calls only: to the user say 'faire connaître' / 'get known' (tofu), 'faire comprendre' / 'get understood' (mofu), 'faire acheter' / 'get bought' (bofu), and the post types in words.

Conversations waiting for a human answer (escalation cards, the app's 'Action requise'): the AI handed over (out of scope, a human requested, a custom stop condition), the prospect replied while the AI that answers was off, the user wrote by hand, the prospect sent a voice note, or a won or closed prospect wrote again. Nothing to approve here, and list_pending_messages does not list them. The machine will NOT answer these; they wait for the user. Read each thread with get_conversation, then help the user craft an answer and send it via reply_to_lead. Covers every campaign by default (pass campaign_id to narrow to one client). total / has_more / next_offset describe escalation_cards; leads_waiting_total is the real number of prospects waiting for the user, of which the 30 most recent are listed.

List the user's prospects (leads found and worked by their automated LinkedIn prospecting). Filterable by status (comma-separated; 'connected' expands to every status where the LinkedIn relationship exists), tag name, or free-text search on name/position. Status meanings: new = found not yet contacted, connection_request = invite sent, connection_accepted = invite accepted, first_contact = first DM sent, discussion = the AI is talking with them, follow_up = being reminded, snoozed = re-engage at a future date, goal_reached = WON, talk_to_human = needs the user, dropped = abandoned. These codes are internal: report a status to the user in plain words in their language ('invitation envoyée', 'en discussion', 'objectif atteint'), never as the raw code. Paginated: `total` is the real number of matching prospects and `has_more` says whether you only saw a slice -- never present one page as the whole base, page with `next_offset` or narrow the filters.

LinkedIn conversations in the user's inbox that MimikFlow (the machine automating their LinkedIn prospecting 24/7) is NOT tracking: threads the user started or handled by hand, outside the automated pipeline. Scans the most recent chats and returns the untracked ones with the contact's name, headline and a chat_id. Use it when the user wants to see their manual conversations or hand one over to the machine; then call adopt_conversation with the chosen chat_id. Read-only: nothing is saved or contacted.

The campaign's media library: the videos, documents, images and links MimikFlow can attach to a message. Each one carries a title, a description (what it is and when it is relevant) and, when the user set one, an explicit rule saying when the AI may send it. Use it whenever a conversation would land better with something to SHOW -- a demo video before a call, a case study after one, the document a prospect asked for -- and to get the `media_id` that attaches a file to a message around a booked meeting (update_meeting_moments). Most users upload one file and never wire it anywhere, so an unused media is worth mentioning once when their conversations show it would help. Files are uploaded in the app; this tool only reads what exists.

Meetings the AI setter booked with prospects in the user's real calendar (the usual conversion goal). Returns prospect, datetime and status. upcoming_only=false includes past meetings. CAVEAT: only meetings stamped through a CONNECTED calendar appear here; a meeting agreed in conversation or self-booked via a link often does not. goal_reached in get_pipeline_stats is the source of truth for won leads -- never conclude 'no meetings booked' from this list alone.

The user's approval inbox: messages MimikFlow's AI drafted for prospects but is holding until the user validates them (validation mode, or a coherence check flagged the draft). A processing item is visible but must not be acted on until MimikFlow finishes or asks for a calendar check. Workflow: read each lead's thread with get_conversation, then approve_pending_message (send as is), edit_and_approve_pending_message (improve then send), or reject_pending_message (never send). Covers every campaign by default (pass campaign_id to narrow to one client), oldest first, and paginated: `total` is the real backlog, `has_more` says a slice is missing -- page with `next_offset` before telling the user their inbox is empty or counting it. Escalation cards (conversations handed to the user, nothing to approve) are NOT listed here: `escalations_waiting` counts them in the same scope, list them with list_escalations, and never tell the user nothing is waiting while it is above 0.

List the user's lead tags (free labels used to segment prospects, e.g. by cohort, event or offer) with the number of leads carrying each. Tags are usable as a filter in list_leads. Complete list, never truncated.

The WhatsApp numbers connected to the user's MimikFlow, and what each one is allowed to do right now. WhatsApp is a second channel on the SAME prospects as LinkedIn: it answers people who write first (a wa.me link, an ad, a QR code), carries the messages around a booked call, fills and converts a live, re-activates an imported base, and can take over when LinkedIn goes silent. Read `quiet_reason` before telling the user anything is broken: a freshly connected number stays SILENT for 24 hours and then writes only a few messages a day for its first two weeks, on purpose. That is WhatsApp's own rule against numbers that behave like machines, not a malfunction and not something to work around -- explain it and say how long is left. `paused` means the number was put on hold automatically because too few of the conversations it opened got an answer; the fix is better targeting and a better first message, never more volume. Cold-outreach and reply ceilings, credits left and whether unknown senders are accepted are reported as they are: none of them is writable from here.

Prepare one editable campaign proposal from the user's website and any business facts they already gave. Use it after get_onboarding_status returns prepare_setup. The result includes the offer, ideal customer, sender identity, and writing style, but saves none of those fields. It deliberately proposes NO exclusion list: every prospect is compared to it and one nobody chose drops the right people in silence, so a precise ideal customer is the answer here, never a counter-list. Show the proposal in plain language and ask the user to approve or correct it before calling save_onboarding_setup. This can read a public website and use MimikFlow's existing bounded onboarding AI; it never activates prospecting or contacts anyone.

Reject a pending draft (it will never be sent). Setter drafts move the lead to talk_to_human. Cold DM (first message) drafts DROP the lead: rejecting one means the prospect leaves the pipeline entirely, not just the wording -- tell the user, and if they only disliked the text, use restore_leads then send_first_message to get a new draft. Follow-up drafts restore the previous status. Also marks an escalation card as handled (it only removes the notification: the lead keeps its status, a won prospect stays won).

Send one message to a prospect AS the user, in their existing conversation. Meant for conversations the AI handed over (talk_to_human) after reading the full thread with get_conversation, and for a quick courtesy message on a WON prospect right after a meeting ('merci pour ce call, je t'envoie un mail') -- a won prospect keeps its won status. Write in the conversation's language and the user's voice, and get the user's OK on the wording before sending: this message goes out immediately. SAFETY: the live LinkedIn thread is refreshed first, then the tool REFUSES to write when we already spoke last and the prospect has not answered since -- that is how a prospect the user answered by hand ends up reminded twice. The refusal returns the real last messages: read them, tell the user what is actually in the thread, and only pass allow_consecutive=true when they confirm they do want a second message in a row. channel: 'linkedin' (default) or 'whatsapp' when the prospect's number is connected and the conversation lives there. WhatsApp goes out under the number's own protections (daily ceiling, spacing, credits, opt-out): a refusal there is a safety rule doing its job, never a bug to work around.

After checking the real calendar for a draft marked calendar_outcome_unknown, record whether the meeting exists. This never sends a message. Use 'booked' only when the event is visible, otherwise use 'not_booked'. The draft then returns to normal review.

Put prospects that were dropped (status 'dropped', abandoned) back into MimikFlow's automated pipeline (MimikFlow is the machine automating the user's LinkedIn prospecting). This is the return path after drop_leads or after rejecting a first-message draft, which also abandons the prospect. Each prospect goes back to the stage its own history justifies: a prospect who already wrote returns to the live conversation and the AI answers on its next run if they spoke last; a prospect we wrote to returns to the reminder sequence; an accepted connection with no message yet waits for its first message, which the machine writes at its normal safe pace (use send_first_message to trigger one now); a pending invitation waits for its answer; anything else starts over. Only works on dropped prospects. Confirm with the user before calling: the machine will write to these people again. Max 50 ids per call (from list_leads with status='dropped').

Give a conversation back to the AI. When MimikFlow (the machine automating the user's LinkedIn prospecting) hands a prospect over to the human -- the user answered by hand on LinkedIn, the AI escalated, or a reply landed while the AI was off -- that conversation stays manual forever: the AI stops answering and the reminders stop. This is the return path. It first syncs the real LinkedIn thread (so the messages the user typed by hand are recorded and stop being read as a new manual takeover), then puts the prospect back under the AI: it answers on its next run when the prospect spoke last, otherwise the reminder sequence resumes. The prospect's history and the notes the user wrote are kept, and the 'needs your attention' cards for that prospect are cleared. Only works on a prospect currently handled by the human. If the user answers by hand again afterwards, the conversation goes back to them, which is intended. Confirm with the user before calling: the AI will write to this prospect again.

Trigger an immediate one-off run of a MimikFlow pipeline instead of waiting for its schedule (e.g. lead_scraper to go find prospects right now, connection_sender to send today's invites). All daily LinkedIn safety limits still apply, so 'now' never means 'more'. Not available for setter (it runs whenever a prospect replies) and tracking. To the user, describe the action in product words in their language ('je relance la recherche de prospects'), never with the internal pipeline id.

Save only the campaign proposal the user explicitly approved. Call this after prepare_onboarding_setup, once the user has seen the offer, ideal customer and writing style and said they are correct. Pass approved=true plus only the accepted fields. An ideal customer that makes a need, an intention or a buying signal required is refused with the lines to rewrite: those are bonus signals, never requirements. The save is idempotent, preserves every omitted value and every existing campaign, and never activates prospecting or sends anything. After saving, follow the returned next_action: connect LinkedIn if needed, otherwise preview a small set with search_prospects without importing it.

Full-text search across every prospect conversation (e.g. 'pricing', 'not interested', a competitor name). Returns matching leads with the matched snippets. Scans the 200 most recent matching messages only: `truncated` true means older matches exist beyond what is shown, so present the results as examples, never as an exhaustive count.

MimikFlow's official help center content. Use it to answer ANY how-to or support question about MimikFlow (setup, features, plans, LinkedIn safety, billing, calendar, broadcasts, MCP...) with accurate product answers instead of guessing. Without arguments: the catalog of topics with their questions. With topic_id: that topic's full content. With query: every topic matching the text. Base your answer on this content and answer in the user's language.

Preview-search LinkedIn prospects matching a free-text query (e.g. 'expert comptable Bordeaux' or 'SaaS founder Paris hiring SDR'). Returns candidate profiles WITHOUT contacting or saving anything, marking those already in the campaign. Review with the user, then pass the chosen URLs to add_prospects_to_campaign.

Write and send the first message to ONE accepted LinkedIn connection now, instead of waiting for MimikFlow's next automatic run (MimikFlow is the machine automating the user's LinkedIn prospecting; it normally writes first messages by itself, at a safe pace). Only works on a prospect at status 'connection_accepted' with no message yet; a dropped prospect needs restore_leads first, a prospect already in conversation gets reply_to_lead. The message is written by the AI from the campaign's settings and this prospect's profile, in the prospect's language. In validation mode the message is NOT sent: it lands as a draft the user approves with approve_pending_message, and the answer says so. Every call counts against the account's daily first-message limit, which can never be exceeded from here: call it for a few chosen prospects, never in a loop over a list. Confirm with the user before calling.

Give the machine fresh news from the user (a new case study, a new result, a new offer detail). For the next 30 days, MimikFlow swaps ONE reminder per lead with a re-offer message built from this news, a strong way to revive silent prospects. Max 600 chars, strictly factual (the AI must never invent numbers). Empty text clears it.

SHOW the user their MimikFlow cockpit (the machine that runs their LinkedIn prospecting: each step on or off, prospects at every stage, what waits for them, their results). In assistants that display apps it opens an interactive panel where the user can act; elsewhere it answers the same data as get_account_overview. Call it ONLY when the user asks to see or open their cockpit, dashboard or MimikFlow ('montre-moi mon cockpit', 'ouvre MimikFlow', 'show me my dashboard'), and once per request. To read the account for your own reasoning, call get_account_overview instead: it shows the user nothing.

Schedule a future re-engagement for a lead. Two uses: (1) the prospect asked to be re-contacted later ('come back next month') -- the lead is snoozed and MimikFlow wakes it at the date with one warm re-engagement message, then resumes its normal cadence; (2) after a meeting happened (lead already won / goal_reached) -- the lead KEEPS its won status and gets ONE follow-up at the date, with no automatic cadence after it. Pass the optional `note` to give the AI real context for that message (summary of the meeting or call, the offer made, the agreed next step): the follow-up will reference it naturally instead of shipping a generic re-engagement. callback_at is an ISO date clamped to 1-365 days from now; null cancels the schedule and clears the note.

Attach a tag (free label used to segment prospects) to a list of leads, creating the tag if it does not exist yet. Idempotent: already-tagged leads are unaffected. Useful to mark a cohort before filtering or reporting on it later.

Turn one of MimikFlow's automated pipelines on or off for a campaign. The prospecting chain: lead_scraper ('la recherche de prospects', finds new prospects that match the ICP), connection_sender ('les demandes de connexion', sends LinkedIn invites), cold_dm_sender ('le premier message', first personalized message), follow_up ('les relances', spaced reminders), setter ('l'IA qui répond aux prospects'). Extra lead sources: engagement_scraper ('les réactions à tes posts', people who like/comment the user's posts), profile_views ('les visiteurs de profil'), invite_manager ('les invitations reçues'), first_degree_scanner ('tes relations de 1er degré', existing connections). These ids are tool-call vocabulary only: to the user, always name the step by its product name in their language, never by its id. Changes apply within minutes; use run_pipeline_now for an immediate run. Warn the user before turning OFF part of the chain: leads then pile up unprocessed at that step.