Context Vigilance
Search/

← Corpus / lossless-monorepo / agent-skill

lossless-monorepo

lossless-monorepo/agent-skills/decile-hub-connector/references/endpoint-inventory

Path
agent-skills/decile-hub-connector/references/endpoint-inventory.md

Decile Hub API v1 — Endpoint Inventory

Exhaustive list of every path in the OpenAPI spec, grouped by tags. Generated from ai-labs/augment-it/clients/humain-vc/inputs/decilehub/202506_decilehub-docs_swagger.yaml (OpenAPI 3.0.1, Decile Hub API v1, 11,970 lines). Format:

METHOD /path — purpose — params — request body — response.

All routes are under the per-tenant base URL https://<tenant>.decilehub.com. Auth: Authorization: <raw-token>. Operations marked [agent-tool] carry x-agent-tool: true in the spec (expose these first in the MCP).

Authentication

  • GET /api/v1/whoami — identify the authenticated token (kind, user, account, account_user.roles, accessible_pipeline_ids). 403 on legacy tokens.

Admin

  • GET /api/v1/admin/accounts — list accounts the calling admin can access. Query search; requires X-Account-Id header; admin token only.

Accounts

  • GET /api/v1/accounts — current account basic info.

AccountUsers

  • GET /api/v1/account_users — list team members (discover assigned_id for ownership). Query page (0-idx, size 100). Pattern A.

ActivityEntries

  • GET /api/v1/activity_entries — activity feed, newest first. Many filters (subject_type, subject_id, event, user_id, created_after/before, page_token, per_page≤100). Pattern C (next_page_token, has_more).
  • GET /api/v1/activity_entries/{id} — single entry + full entryable body → ActivityEntryDetail.

Entities (funds, SPVs, holdings, management companies, general partnerships)

  • GET /api/v1/entities — list firm-admin entities. Query type[] (fund/spv/management_company/general_partnership/holding), active, page (1-idx), per_page. Pattern B.
  • POST /api/v1/entities — create entity (+ Organization + FundDetail). Body EntityCreateRequest. Not idempotent. Requires full-access account admin.
  • GET /api/v1/entities/{id} — show entity + details (+ optional include=calculations, window=now|lifetime).
  • PATCH /api/v1/entities/{id} — update writable fields (kind immutable). Body EntityUpdateRequest. 409 if org shared with siblings.

Capital Accounts

  • GET /api/v1/entities/{entity_id}/capital_accounts — identity-only list. Query active, partner_type, page, per_page. Pattern B.
  • GET /api/v1/capital_accounts/{id} — identity + commitments + contacts + entity (+ include=transfers).
  • GET /api/v1/capital_accounts/{id}/calculations — single-CA period figures. Query period_start, period_end.
  • GET /api/v1/entities/{entity_id}/capital_accounts/calculations — bulk calc + totals.

Journal Entries (GL)

  • GET /api/v1/entities/{entity_id}/journal_entries — list GL transactions (many filters). Pattern B.
  • POST /api/v1/entities/{entity_id}/journal_entries — create one GL entry. Not idempotent.
  • POST /api/v1/entities/{entity_id}/journal_entries/bulk — atomic batch (1–100). 422 → JournalEntryBulkErrors.
  • GET /api/v1/journal_entries/{id} — show one.

Accounting Accounts (chart of accounts)

  • GET /api/v1/accounting_accounts — list GL codes (filters code, code_prefix, is_1099_eligible, search). Pattern B.
  • GET /api/v1/accounting_accounts/{id} — show one.

Capital Calls

  • GET /api/v1/capital_calls — list. Query entity_id (required), status, period_start/end. Pattern B.
  • GET /api/v1/capital_calls/{id} — header for one call.
  • GET /api/v1/capital_calls/{id}/details — per-LP rows.

Financial Reports

  • GET /api/v1/financial_reports — list report jobs. Pattern A.
  • POST /api/v1/financial_reports — queue async report job (202). report_types enum or all_reports; format xlsx/pdf.
  • GET /api/v1/financial_reports/{id} — job status (poll status_url).

Events

  • GET /api/v1/events — list events (filters title, start_date, end_date, published, access_type). Pattern A.
  • POST /api/v1/events — create event (auto-creates pipeline + questionnaire).
  • GET /api/v1/events/{id} — show event (by id or unique_event_id; include=guests).
  • PUT /api/v1/events/{id} — update event.
  • DELETE /api/v1/events/{id} — delete event + associated data.
  • POST /api/v1/events/{id}/guests — bulk-add guests (creates People). Returns {added, duplicates, errors}.
  • GET /api/v1/events/{id}/guests — list guests w/ RSVP + stage.
  • PATCH /api/v1/events/{id}/rsvp — update guest RSVP (email, status).

Files & Folders

  • GET /api/v1/files — list files (filters name, folder_id, extension, dates). Pattern A.
  • POST /api/v1/files — upload (multipart). file[file], file[attachable_type] (Person/Organization/PipelineProspect), file[attachable_id], folder_id.
  • GET /api/v1/files/{id} — show file.
  • PUT /api/v1/files/{id} — update metadata or upload new version (multipart).
  • GET /api/v1/files/{id}/download — download (binary).
  • POST /api/v1/files/{id}/save_to_folder — save an in-message Attachment into a data-room folder (idempotent). (not [agent-tool])
  • GET /api/v1/folders — list folders.
  • POST /api/v1/folders — create folder (name, parent_id).
  • GET /api/v1/folders/{id} — show folder + children (contents filter).
  • GET /api/v1/folders/search — search folders by name. Query q (required). (not [agent-tool])

Directory — People & Organizations (the CRM core) [agent-tool]

  • GET /api/v1/organizations — list orgs. include (notes,people,referred_by,…), fields, custom_data_points, name, dates, page (0-idx). Pattern A.
  • POST /api/v1/organizations — bulk add orgs (first 100). → {created, duplicates, errors}.
  • POST /api/v1/organization — upsert ONE org (natural key name). → {status, organization_id, changes}.
  • GET /api/v1/organizations/{id} — show org (+ logo = attached_image).
  • POST /api/v1/organizations/{id}/notes — append note (body, context).
  • GET /api/v1/people — list people. include (notes,referred_by,organizations), fields, custom_data_points, first_name, last_name, email, dates, page. Pattern A.
  • POST /api/v1/people — bulk add people (first 100). → {created, duplicates, errors}.
  • POST /api/v1/person — upsert ONE person (natural key email). → {status, person_id, changes}.
  • GET /api/v1/people/{id} — show person (+ picture = attached_image).
  • POST /api/v1/people/{id}/notes — append note.

Tasks

  • GET /api/v1/tasks — list (requires :hub_tasks flag, else 404). Filters status, assignee_id, origin. Pattern A.
  • POST /api/v1/tasks — create user-origin task (title, assignee_id required).
  • GET /api/v1/tasks/{id} — get one.
  • PATCH /api/v1/tasks/{id} — update.
  • POST /api/v1/tasks/{id}/complete — mark completed.

PipelineProspects (CRM deal/relationship rows) [agent-tool]

  • GET /api/v1/pipeline_prospects — list. pipeline_id (required), many filters (stage, tags, contact dates), page. Pattern A. (x-agent-category: context)
  • POST /api/v1/pipeline_prospects — bulk create Person/Org prospects (first 100 each).
  • PATCH /api/v1/pipeline_prospects — batch update (resolve by id/name/email). Partial success → 200 with errors[].
  • PATCH /api/v1/pipeline_prospects/{id} — update one (query pipeline_id required). Returns bare {error: string} on failure.
  • POST /api/v1/pipeline_prospects/{pipeline_prospect_id}/notes — append note.
  • POST /api/v1/pipeline_prospect — upsert ONE prospect (exactly one of person|organization). → {status, pipeline_prospect_id, changes}.
  • GET /api/v1/pipeline_prospects/{id}/action_executions — list queued/run pipeline actions (tree).
  • POST /api/v1/pipeline_prospects/{id}/action_executions/{id}/execute — execute one action.
  • POST /api/v1/pipeline_prospects/{id}/action_executions/{id}/preview — preview a pending send_email action.

Pipelines

  • GET /api/v1/pipelines — list active pipelines (filter kind). (x-agent-category: context)
  • GET /api/v1/pipelines/{id} — pipeline + stages.
  • GET /api/v1/pipelines/{id}/metrics — current + 7-day-ago snapshot.
  • POST /api/v1/pipelines/{pipeline_id}/data_points — create a custom data point (Variable) for the pipeline TYPE (account-wide fan-out; account admin). Format enum: string, paragraph, url, number, date, date_us, currency_us, currency_eu, percent, percent_rounded, select, file_url.

ResearchInvestors (shared Decile Research LP roster) (not [agent-tool])

  • GET /api/v1/research/investors — list shared LP roster.
  • POST /api/v1/research/investors/copy_to_pipeline — copy entries to a stage (202).

Deal Shares (network feed)

  • POST /api/v1/deals/share — create/update deal share (idempotent per org). Required: organization_id, company_name, the_bet, referring_manager_name, referring_manager_email.
  • GET /api/v1/deals/shares — list own deal shares. Pattern C.
  • GET /api/v1/deals/shares/{id} — get one.
  • DELETE /api/v1/deals/shares/{id} — delete one.
  • POST /api/v1/deals/shares/{id}/copy_to_pipeline — copy shared deal into a stage (202).
  • POST /api/v1/deals/shares/ai_auto_fill — AI auto-fill enrichment (synchronous LLM).

Deal Memos

  • GET /api/v1/deal_memos — list (filters organization_id, stage). Pattern C.
  • POST /api/v1/deal_memos — start a memo (organization_id required).
  • GET /api/v1/deal_memos/{id} — full memo.
  • PATCH /api/v1/deal_memos/{id} — update (422 if closed).
  • POST /api/v1/deal_memos/{id}/submit_for_review — submit for review.
  • POST /api/v1/deal_memos/{id}/close — close with close_type (approve/pass).

PortfolioCompanies

  • GET /api/v1/portfolio_companies — list (each = fund×org pair). Filters fund_id, organization_id. Pattern C.
  • GET /api/v1/portfolio_companies/{id} — get one.
  • GET /api/v1/portfolio_companies/{portfolio_company_id}/investments — list investment tranches.
  • GET /api/v1/portfolio_company_investments/{id} — get one tranche.

Pacts / Lpas

  • POST /api/v1/pacts/send_pact — send PACT email to a Person (person_id, pipeline_id).
  • POST /api/v1/lpas/send_lpa — send LPA email to a Person (moves to Closing, creates capital account).

Emails

  • GET /api/v1/email_templates — list templates. Pattern A.
  • GET /api/v1/email_templates/{id} — full template + variables: [{name, required}] schema.
  • POST /api/v1/emails/preview — preview templated/freeform email (exactly one recipient field).
  • POST /api/v1/emails — send/schedule/test-send. Requires confirm: true (422 confirmation_required otherwise).

Variables (merge-tag / data-point picker) (not [agent-tool])

  • GET /api/v1/variables — variable picker for pipeline_id (required). Pattern A.
  • GET /api/v1/variables/{id} — show one variable (query pipeline_id required).

Base (Decile Base community)

  • GET /api/v1/base/inbox — inbox feed. Pattern D (meta).
  • GET /api/v1/base/channels — list channels.
  • GET /api/v1/base/channels/{id}/posts — posts in a channel.
  • POST /api/v1/base/posts — create post (Markdown).
  • GET /api/v1/base/posts/{id} — show post + replies.
  • POST /api/v1/base/replies — reply to a post.
  • POST /api/v1/base/search — full-text search posts + articles.
  • GET /api/v1/pipeline_prospects/{pipeline_prospect_id}/attachments — list a prospect’s files; merges source: "direct" (CRM attachments on the org/person) and source: "questionnaire" (founder’s deal-intake uploads). 403 unless the pipeline is investment-type. Not in the 202506 swagger — live docs only.
  • POST /api/v1/pipeline_prospects/{pipeline_prospect_id}/attachments — attach directly (multipart: attachment[file], optional attachment[name]). No data-room copy. Needs edit_prospects?. Live docs only.
  • DELETE /api/v1/pipeline_prospects/{pipeline_prospect_id}/attachments/{attachment_id} — delete a direct attachment only. Live docs only.
  • GET /api/v1/pipeline_prospects/{pipeline_prospect_id}/attachments/{signed_id}/download — download by signed blob id. Live docs only.
  • GET /api/v1/base/attachments/{id} — download a blob (binary / 302 if >50MB).

agent_platform (not [agent-tool])

  • POST /api/v1/agent_platform/file_search — semantic search over account corpus (q, limit).

Key schemas (field lists)

Full field lists for the CRM-core resources are in the SKILL.md “Pushing data” + “Mapping” sections. The named component schemas worth knowing: PortfolioCompany, PortfolioCompanyInvestment(Detail), Entity/EntityDetails (funds), CapitalAccountIdentity/Commitment/CapitalAccountContact, Task, note_envelope, ActivityEntry(Detail), DealShare, variable, attached_image, associated_person, address, ErrorResponse. Person and Organization have NO standalone schema — their stored fields are dynamic (data / custom_data_points jsonb); the write fields are documented inline (see SKILL.md). Read the spec for exact field-level detail.