{"id":"io.agent4/agent4-tenant","slug":"io-agent4-agent4-tenant","title":"agent4.io","description":"Build and run grounded business agents over MCP: agents, knowledge bases, skills, Storylines.","version":"1.4.0","publisher":{"kind":"domain","id":"agent4.io","label":"agent4.io","url":"https://agent4.io"},"website":"https://agent4.io","repository":"","remotes":[{"type":"streamable-http","url":"https://api.agent4.io/v1/mcp","headers":[{"description":"Your agent4.io tenant API key (console → Settings → Security; shown once at creation).","isRequired":true,"isSecret":true,"name":"X-API-Key"}]}],"packages":[],"published":"2026-07-30","updated":"2026-07-30","status":"online","auth":"","latency_ms":1187,"checked":"2026-11-19","uptime":100,"checks":3,"tools":[{"name":"list_agents","title":"","description":"List all agents in this tenant (name + published state).","read_only":false},{"name":"get_agent","title":"","description":"Get one agent's full configuration (soul/task/tools/skills/knowledge_bases/model/params).","read_only":false},{"name":"create_agent","title":"","description":"Create an agent.\n\n    `name` is lower-cased on save — it is a key, not a display name.\n\n    soul = persona/identity; task = duties and boundaries; tools = tool-name whitelist (check\n    list_tools first; the system tools time/geo/weather **and `compute_chart`** are\n    **auto-enabled by default**, no need to list them — but note that passing `tools` at all\n    replaces that default list, so includ","read_only":false},{"name":"update_agent","title":"","description":"Update **part** of an agent — fields you don't pass stay as they are.\n\n    ⚠️ List-field semantics, don't mix them up: `tools=[...]` / `skills=[...]` /\n    `knowledge_bases=[...]` are **full-list replacement** — items you didn't read first get pushed\n    out. To \"add one / remove one\", use `add_tools` / `remove_tools` (and add_/remove_skills,\n    add_/remove_knowledge_bases): incremental, idempote","read_only":false},{"name":"list_tools","title":"","description":"List tool names assignable to agents, with their descriptions (incl. the tenant's connected MCP tools).","read_only":false},{"name":"list_mcp_servers","title":"","description":"List the MCP servers connected in this tenant's tool library (read-only; secret fields omitted).","read_only":false},{"name":"create_share","title":"","description":"Create a **share / integration entry point** for an agent — this is how end users actually reach it.\n\n    **`published=True` only means \"visible\", not \"reachable\"**: for end users to talk to the agent\n    you must create a share. The response carries a directly openable chat link\n    (`{public_base}/s/<token>`) and the website embed URL (`{public_base}/embed/<token>`).\n    For a website widget, pa","read_only":false},{"name":"list_shares","title":"","description":"List an agent's shares (token / label / disabled / chat link / embed URL / appearance config).\n\n    When the top-level `pretty_url` is non-empty it is the preferred link to give humans\n    (see create_share).","read_only":false},{"name":"configure_share","title":"","description":"Configure this share — its name and its **appearance** — fields you don't pass stay as they\n    are (server-side partial merge).\n\n    - `label`: the name at the top of the chat page, and the browser tab's title. **Renaming a\n      share is this call**; you never need to create a replacement and delete the old one. That\n      would mint a new token, so the link already embedded in the client's site","read_only":false},{"name":"set_pwa_branding","title":"","description":"Configure a specific **agent's** PWA branding and install experience (applies to that\n    agent's `/s/` standalone chat pages — what gets installed to the home screen is one agent's\n    entry page, so each agent is its own app).\n\n    - `agent`: the agent name whose install branding to configure (required).\n    - `icon_source_url`: public URL of one master image (PNG/JPG/WebP ≥192×192, ideally a sq","read_only":false},{"name":"set_custom_domain","title":"","description":"Bind the client's own domain to this tenant's chat pages: `https://chat.client.com/` serves\n    the tenant's branded page (or a specific agent's chat), the address bar keeps the client's\n    domain, and the TLS certificate is issued automatically.\n\n    **Have the client configure DNS first**: add a CNAME at their DNS provider pointing to the\n    `cname_target` in the response (`endpoint.agent4.io`","read_only":false},{"name":"list_skills","title":"","description":"List this tenant's skills.","read_only":false},{"name":"get_skill","title":"","description":"Get one skill's full content (including the complete instructions text).","read_only":false},{"name":"create_skill","title":"","description":"Create a skill. description = \"when to use\" (goes into the system prompt, keep it short);\n    instructions = the detailed guide (fetched on demand via load_skill).\n\n    The paradigm (important): instructions are visible only AFTER the model calls load_skill — if\n    calling some tool is **mandatory** (\"user gives a phone number → must call save_contact\"), the\n    trigger must be written into descr","read_only":false},{"name":"update_skill","title":"","description":"Update part of a skill — fields you don't pass stay as they are (server-side PATCH merge,\n    no GET+PUT round-trip).\n\n    ⚠️ `tools=[...]` is **full-list replacement**; to add/remove one use `add_tools`/`remove_tools`\n    (incremental, idempotent). Tools bound to a skill take effect automatically at chat time on any\n    agent the skill is attached to — no need to also add them to the agent's tool","read_only":false},{"name":"test_skill_trigger","title":"","description":"Reality-check whether your prompts actually trigger tool calls (dry-run) — run this after\n    writing/changing a skill instead of counting corpses in production.\n\n    Replays your messages N times against the **production** system-prompt assembly, tool schemas\n    and this tenant's actual model routing, capturing only the model's tool-call decision: **tool\n    side effects are NOT executed**, no s","read_only":false},{"name":"list_knowledge_bases","title":"","description":"List this tenant's knowledge bases (with doc/chunk counts and relevance cutoff).","read_only":false},{"name":"get_knowledge_base","title":"","description":"Get one knowledge base's configuration (including the full instructions text shown to the model).","read_only":false},{"name":"create_knowledge_base","title":"","description":"Create a knowledge base.\n\n    **Write `description` — it is what routes questions to this base.** Before each answer the\n    platform reads every attached base's `description` and decides which ones this question needs;\n    a base whose description is vague or blank gets searched when it should not, or missed when it\n    should. One line, in the words a visitor would use, naming the subject **and ","read_only":false},{"name":"update_knowledge_base","title":"","description":"Update part of a knowledge base — fields you don't pass stay as they are (server-side PATCH merge).","read_only":false},{"name":"add_knowledge_text","title":"","description":"Add a **text** document to a knowledge base (chunking + embedding happen synchronously;\n    returns the real chunk count).\n\n    For **local files** (pdf/docx…) see `add_knowledge_file`.\n    ","read_only":false},{"name":"add_knowledge_file","title":"","description":"Add a local file's content to a knowledge base (txt/md/html/pdf/docx).\n\n    **This MCP runs on the platform server and cannot read paths on YOUR machine.** For text\n    files, read the content yourself and call `add_knowledge_text`; for binaries (pdf/docx),\n    upload via the console, or curl `/knowledge-bases/{name}/documents/upload`.\n\n    **To ingest a whole folder, zip it and send that one endp","read_only":false},{"name":"search_knowledge_base","title":"","description":"Search this knowledge base; returns matching chunks with distances.\n\n    Uses **exactly the same retrieval path as real conversations** — use it to validate the cutoff\n    and chunking: an empty result means this question will be judged \"not covered\" and the agent\n    will answer \"not in the knowledge base\".\n    ","read_only":false},{"name":"build_knowledge_index","title":"","description":"Build (or rebuild) the structured index for a knowledge base — the second leg beside vector search.\n\n    Vector search answers \"what does this passage say\". It **cannot count, filter numerically or\n    aggregate**, so \"how many documents\", \"which ones are between 1000 and 2000 words\", \"how many\n    per category\" are not answered badly — they are structurally unanswerable. This builds a small\n    p","read_only":false},{"name":"get_knowledge_index","title":"","description":"Show the structured index profile: per column coverage, type, numeric range, top values, warnings.\n\n    **The warnings are the point.** On a real 4,500-document catalogue this surfaced three data\n    problems nobody knew about: 1,058 documents with `page_count` 0, 505 with `reading_level` 0,\n    493 with `word_count` 0 — those are not zeros, they are missing values recorded as 0, and they\n    sile","read_only":false},{"name":"patch_knowledge_index","title":"","description":"Change the structured index schema from one sentence of plain language.\n\n    Example requests: \"also track the author so users can find other books by them\", \"I want to\n    filter by illustrator\", \"drop the cover link\". The model may only emit add/modify/remove/refused\n    — it cannot rewrite the table, so it has no way to damage columns the user did not mention.\n\n    **Data that is not in the doc","read_only":false},{"name":"improve_prompt","title":"","description":"Rewrite a field's text to work better as an AI system prompt; returns the improved text.\n\n    field ∈ persona | task | greeting | storyline_task | storyline_opening\n            | storyline_ai_trigger | storyline_ai_criteria\n            | skill_description | skill_instructions | kb_description | kb_instructions\n    — determines the rewrite target (a Task should read like commands, a KB description ","read_only":false},{"name":"diagnose_prompt","title":"","description":"Read a field and return a list of **specific** improvement suggestions (string array).\n\n    field: same values as improve_prompt. current = the text to diagnose (required — an empty text\n    has nothing to diagnose). Typical flow: diagnose_prompt first, pick the suggestions you want,\n    then fold them into improve_prompt's hint.\n    ","read_only":false},{"name":"test_planner_trigger","title":"","description":"Dry-run the Dynamic Planner's escalation judge: would these messages be offered a\n    step-by-step plan? Use it after editing the agent's `task`, since the task text is the\n    **business-domain gate** — a vague task lets out-of-domain requests through, a narrow one\n    turns real customers away.\n\n    messages: the user's turns in order; the LAST one is treated as the current turn and the rest\n   ","read_only":false},{"name":"preview_plan","title":"","description":"Generate a plan for a goal and return it for review **without persisting anything** — no\n    temporary storyline, no enrollment, no user affected. Use it to inspect plan quality before\n    trusting the feature in production, or to see how the agent's `task` constrains the steps.\n\n    Returns {plan, backend, warnings}. `warnings` flags machine-detectable smells, notably checklist\n    items that are","read_only":false},{"name":"usage_stats","title":"","description":"Usage statistics. group ∈ total | agent | user | space | day | model.\n\n    Metadata only — token counts and event counts, never any conversation content.\n    Use group=\"user\" to see who uses the most, group=\"agent\" for the busiest agent.\n    ","read_only":false},{"name":"tenant_info","title":"","description":"This tenant's basic info and quota (plus custom-domain binding status, if any).","read_only":false},{"name":"search_agent4_docs","title":"","description":"Search agent4.io's own product docs — concepts (what a thing *is*) and cookbook recipes (which\n    tool to *call*). Use when you're unsure how an agent4.io feature works, what a term means, or how to\n    build something on the platform. Returns top passages with title, layer (\"concept\"|\"ops\"), url and a\n    snippet. Public content; no tenant data involved.","read_only":false},{"name":"list_end_users","title":"","description":"List this tenant's **end users** (the people chatting with your agents) — roster only,\n    never any conversation content.\n\n    Each user carries: display name, login method (email / oauth:* / proxy), email (if any),\n    space/session/document counts, last-24h token usage, and a `console_url` (click through for\n    detail). q filters by name or external id; sort ∈ recent (default) | tokens24h (des","read_only":false},{"name":"get_end_user","title":"","description":"Get one end user's detail: basic profile (name, contact email/phone, city, timezone) + their\n    spaces.\n\n    Never any conversation content. Carries a `console_url` to the user's detail page.","read_only":false},{"name":"list_user_sessions","title":"","description":"List one end user's **sessions** — metadata only: agent, title (AI summary), message count,\n    tokens, timestamps.\n\n    **Message bodies are not returned.** Transcripts are rendered in the console: each session\n    carries a `console_url` that **deep-links straight to that conversation, auto-opened** (no\n    digging through the detail page when there are many), plus `user_console_url` to the user","read_only":false},{"name":"list_page_contexts","title":"","description":"List all page playbooks of this tenant (with match rules, greeting mode, position).","read_only":false},{"name":"upsert_page_context","title":"","description":"Create or fully replace a page playbook.\n\n    `context` is the page background the **server** injects for the agent — write \"who lands on\n    this page, what they are deciding, what they usually worry about\"; do NOT write facts like\n    prices or quotas (those belong in a knowledge base, which also ranks higher in authority).\n\n    `url_pattern` is a glob (`*/pricing`, `*/solutions/*`), matches the","read_only":false},{"name":"resolve_page_context","title":"","description":"Given a URL (or key), show which playbook it resolves to — always verify after writing a\n    match rule.\n\n    Globs make it easy to write rules that \"look right but never match\" (one `*` short, one path\n    level too many) — and in production the only symptom is visitors silently getting the default\n    playbook, with no error anywhere.\n    ","read_only":false},{"name":"page_context_stats","title":"","description":"Open counts and suggested-question click counts per playbook — find copy nobody clicks.","read_only":false},{"name":"list_storylines","title":"","description":"List this tenant's storylines (optionally filtered by agent). Includes state, version, default flag.","read_only":false},{"name":"get_storyline","title":"","description":"Get one storyline's full definition (whole graph + profile_schema + top-level fields).","read_only":false},{"name":"create_storyline","title":"","description":"Create a storyline draft. After creating, self-check with validate_storyline, then\n    publish_storyline.\n\n    user_visibility — what the end user sees of their own run: \"invisible\" (default, no UI),\n    \"named\" (a banner with the storyline name only), \"trail\" (banner + read-only view where\n    untaken branches and future steps are redacted grey blocks), \"full\" (banner with step x/y +\n    full rea","read_only":false},{"name":"update_storyline","title":"","description":"Update **part** of a storyline draft — fields you don't pass stay as they are (GET-then-merge\n    on top of a full PUT underneath).\n\n    concurrency: \"user\" = progress belongs to the person (shared across sessions); \"session\" =\n    progress belongs to the case (one run per session). Changes affect only **future** enrolments;\n    in-flight runs are not migrated.\n\n    ⚠️ Especially `graph`: not pass","read_only":false},{"name":"validate_storyline","title":"","description":"Validate the storyline graph (entry / dead ends / unreachable / dangling / cross-line targets\n    / rule dimensions). Returns {ok, errors}. Must pass before publishing.","read_only":false},{"name":"publish_storyline","title":"","description":"Publish a storyline: validates first (blocking errors → 422), then freezes an immutable\n    version and marks it published.","read_only":false},{"name":"unpublish_storyline","title":"","description":"Unpublish (archive): no more auto-enrolment; in-flight users silently degrade to the plain\n    agent on their next turn (progress kept — re-publishing resumes it).","read_only":false},{"name":"clone_storyline","title":"","description":"Clone a storyline into a new draft (new key, unpublished, doesn't claim default, no chain\n    pointer) for tweaking.","read_only":false},{"name":"export_storyline","title":"","description":"Export a portable storyline JSON (schema_version + definition; no tenant/version/id) for\n    backup, migration or re-import.","read_only":false},{"name":"import_storyline","title":"","description":"Import an exported storyline JSON as a draft. **Artifacts like \"generate a storyline from a\n    novel / source material\" land here.**\n\n    payload = the structure returned by export_storyline (schema_version + graph + profile_schema\n    + top-level fields). Version-aware: payloads newer than the system supports are rejected.\n    Always creates new, never claims default, key collisions auto-rename.","read_only":false}],"server_info":{"name":"agent4-io","version":""},"category":"research","tags":["productivity","web-search","news"],"kind":"remote","source":"registry","submitted":{},"first_party":true,"domain_rank":null,"score":100,"indexable":true,"first_seen":"2026-07-30","stars":0,"badges":[],"sponsor":null,"page":"https://agentsindex.org/agents/io-agent4-agent4-tenant/"}