Atisbo MCP Server

Broad analytical query over claims, solutions, or activity: time filters, grouping, metrics, ready-to-render chart artifact. Use FIRST for analytical questions spanning entities or time windows; for single-entity drill-downs atisbo_lookup mode=claim or atisbo_lookup mode=node is enough.

Capture feedback or knowledge. Modes: signal; signal_batch (idempotent by client_request_id); knowledge; update_artefact (full replace); patch_artefact (preferred—get content_hash with atisbo_lookup mode=doc_patch_context); add_artefact_evidence.

Compact facade for WHERE Atisbo gets its signals — it CONNECTS sources, not just describes them. Modes: list (connections + live status), catalog (ingestible sources; query filters), create (returns the endpoint), configure, pause, resume, sync_now (pull sources only). Call list before any mutation. THE CATALOG IS NOT THE BOUNDARY: almost anything connects — push app (composio_catalog), daily-pull app (composio_actions, ~1,000 apps), raw webhook/email. Never answer "Atisbo cannot ingest that" from an empty search; call catalog and read `fallback`. Atisbo only READS a connected system, never writes. create returns an endpoint + signing secret shown EXACTLY ONCE — relay both to the human immediately.

PM decision facade. CURATE: evidence review/grouping, deal-breakers. EXECUTE: create, update, dispatch or delete Solutions, record Outcomes. DOCUMENT/GOVERN: living docs, coverage, schedules, knowledge, strategy. log_decision is only for external decisions; mutations log themselves. Writes require UUIDs. mode=declare_pm_exception marks a CLAIM as needing a human decision (SEN-180).

Find or open Atisbo entities. work_queue returns an external agent action_queue, completion, Monitoring and pm_exceptions. outcome_review exposes post-launch evidence; implementation_context loads product intent. Mutate with atisbo_decide or atisbo_map.

Compact facade for map work: create/apply maps, generate/edit nodes and edges, anchor claims/snippets, bind metrics, manage claim-snipe maps, regroup claims (merge/split — mode=regroup). Prefer mode=apply for full MECE categorization. TWO WAYS TO PUT A CLAIM ON A NODE, not interchangeable: mode=anchor sets the TAXONOMY category (claims.auto_node_id — one per claim, what the Board groups by), refusing non-taxonomy maps; mode=link_claims adds membership in a STRUCTURAL map (KPI tree, opportunity tree) while the claim keeps its category, refusing taxonomy maps.

Workspace orientation. Modes: snapshot (overview), whats_new (delta), strategy, integrations. Continue the user's existing workflow; never create a parallel backlog. Living docs cite snippet:// refs: open cited Signals with atisbo_lookup mode=open, use claim depth=work for the full corpus, and atisbo_analyze for aggregates. Edit via atisbo_lookup mode=doc_patch_context then atisbo_capture mode=patch_artefact.

Diagnose Atisbo, including "I connected a source and nothing appeared", "I sent text and no opportunity formed", or "my data is missing". Checks integrations, ingestion, clustering, gates and key scope, then returns a playbook. share_report files only diagnostics plus your text—never feedback, customer names or document bodies. Use atisbo_lookup/atisbo_analyze for backlog.