MCP Server
agent4.io
io.agent4/agent4-tenant
What this MCP does
Manages grounded business agents, knowledge bases, skills, connected MCP tools, public shares, and structured interactive storylines.
Capabilities
Tools
50
tools
add_knowledge_file
Add a local file's content to a knowledge base (txt/md/html/pdf/docx).
**This MCP runs on the platform server and cannot read paths on YOUR machine.** For text
files, read the content yourself and call `add_knowledge_text`; for binaries (pdf/docx),
upload via the console, or curl `/knowledge-bases/{name}/documents/upload`.
**To ingest a whole folder, zip it and send that one endpoint** â subdirectories are walked,
md/txt/pdf/html/docx are ingested one by one, and document names use the in-archive relative
path (`2025/rates.pdf` vs `2026/rates.pdf` therefore never collide). Images, `.DS_Store` etc.
inside the zip are ignored; if the archive contains an encrypted, corrupt or abnormally
high-compression file, **the whole archive is rejected** naming the offending entry.
Input schema
{'type': 'object', 'title': 'add_knowledge_fileArguments', 'required': ['kb_name', 'path'], 'properties': {'path': {'type': 'string', 'title': 'Path'}, 'title': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Title', 'default': None}, 'kb_name': {'type': 'string', 'title': 'Kb Name'}}}
Output schema
{'type': 'object', 'title': 'add_knowledge_fileOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
add_knowledge_text
Add a **text** document to a knowledge base (chunking + embedding happen synchronously;
returns the real chunk count).
For **local files** (pdf/docxâ¦) see `add_knowledge_file`.
Input schema
{'type': 'object', 'title': 'add_knowledge_textArguments', 'required': ['kb_name', 'title', 'content'], 'properties': {'title': {'type': 'string', 'title': 'Title'}, 'content': {'type': 'string', 'title': 'Content'}, 'kb_name': {'type': 'string', 'title': 'Kb Name'}}}
Output schema
{'type': 'object', 'title': 'add_knowledge_textOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
build_knowledge_index
Build (or rebuild) the structured index for a knowledge base â the second leg beside vector search.
Vector search answers "what does this passage say". It **cannot count, filter numerically or
aggregate**, so "how many documents", "which ones are between 1000 and 2000 words", "how many
per category" are not answered badly â they are structurally unanswerable. This builds a small
per-KB table from whatever structured header the documents share, which the agent can then query
with SQL via `query_knowledge_table`.
Only worth it when the documents share a machine-readable header (a metadata table, YAML front
matter, `Field: value` lines). **Prose gets declined, and that is the right answer** â a table
of unique values makes statistics meaningless.
`roles` names the fields that must be extracted **exactly** and never paraphrased. Use it when
the answer has to quote something the model must not invent:
- `identity` â what to call the item (book title, drug name, product name)
- `link` â where to send the user
- `image` â what to show the user
- `code` â the unique identifier
Which link is "the" link is a business fact the data does not state â only the customer knows.
A declared role that cannot be found comes back in `roles.unresolved` **with candidate field
names**: ask the user which one it is, do not guess.
**Read `dropped` in the report and tell the user about it.** A column that was thrown out
(coverage too low, two columns holding identical values) is invisible in later query results â
the model simply works around it â so this report is the only place it is ever mentioned.
Input schema
{'type': 'object', 'title': 'build_knowledge_indexArguments', 'required': ['name'], 'properties': {'name': {'type': 'string', 'title': 'Name'}, 'roles': {'anyOf': [{'type': 'array', 'items': {'type': 'string'}}, {'type': 'null'}], 'title': 'Roles', 'default': None}}}
Output schema
{'type': 'object', 'title': 'build_knowledge_indexOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
clone_storyline
Clone a storyline into a new draft (new key, unpublished, doesn't claim default, no chain
pointer) for tweaking.
Input schema
{'type': 'object', 'title': 'clone_storylineArguments', 'required': ['storyline_id'], 'properties': {'storyline_id': {'type': 'string', 'title': 'Storyline Id'}}}
Output schema
{'type': 'object', 'title': 'clone_storylineOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
configure_share
Configure this share â its name and its **appearance** â fields you don't pass stay as they
are (server-side partial merge).
- `label`: the name at the top of the chat page, and the browser tab's title. **Renaming a
share is this call**; you never need to create a replacement and delete the old one. That
would mint a new token, so the link already embedded in the client's site would stop working
â the visible name and the address are not the same thing.
- `theme_color`: the brand colour, `#RGB` or `#RRGGBB`. **This one colour is all you give** â
its foregrounds are derived on save.
- `highlight_color`: **only if the brand actually has a second colour.** Most don't; leave it
out and nothing changes anywhere. When set, it is used where something should be visible
without competing with the primary action: the second series in a chart, and citation
markers. Buttons, the current storyline step and user bubbles stay on `theme_color` â two
colours fighting over the same element is how a two-colour palette goes wrong. Foregrounds
are derived from it the same way.
Text/foreground colours are derived server-side to WCAG contrast and stored (a light brand
colour automatically gets dark text instead of white) â **do not compute a palette yourself
and do not try to set text colours**; they would be overridden.
- `logo_url`: absolute URL. The header scales by height, any aspect ratio fits; **the collapsed
launcher bubble is square**, so only near-square images (aspect 0.74â1.35) are used on the
bubble â a wide wordmark falls back to the platform icon there. Want your mark in both
places? Provide a square variant too.
- `input_mode`: `text` (text box, default) or `voice` (opens in push-to-talk). Requires the
platform's voice backend.
- `launcher`: hover tooltip text for the bubble. `theme`: `light` / `dark` / empty = follow the
visitor.
- `custom_css`: the escape hatch for strict brand guidelines â the injected stylesheet can
override our CSS variables (`--acc` / `--acc-fg` / `--acc-text-l` / `--acc-text-d` / `--bg` /
`--surface` / `--text` / `--border`); write light values under `:root{â¦}` and dark under
`html[data-theme="dark"]{â¦}`. **Prefer theme_color**: it carries a readability guarantee,
with custom_css the contrast is on you.
Get the token from `list_shares` first. Full examples in the Cookbook's configure-branding.
Input schema
{'type': 'object', 'title': 'configure_shareArguments', 'required': ['agent_name', 'token'], 'properties': {'label': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Label', 'default': None}, 'theme': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Theme', 'default': None}, 'token': {'type': 'string', 'title': 'Token'}, 'launcher': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Launcher', 'default': None}, 'logo_url': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Logo Url', 'default': None}, 'agent_name': {'type': 'string', 'title': 'Agent Name'}, 'custom_css': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Custom Css', 'default': None}, 'input_mode': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Input Mode', 'default': None}, 'theme_color': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Theme Color', 'default': None}, 'highlight_color': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Highlight Color', 'default': None}}}
Output schema
{'type': 'object', 'title': 'configure_shareOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
create_agent
Create an agent.
`name` is lower-cased on save â it is a key, not a display name.
soul = persona/identity; task = duties and boundaries; tools = tool-name whitelist (check
list_tools first; the system tools time/geo/weather **and `compute_chart`** are
**auto-enabled by default**, no need to list them â but note that passing `tools` at all
replaces that default list, so include them yourself if you pass any);
knowledge_bases = attached knowledge base names (once attached, retrieval is injected
automatically every turn).
`ask_forms` (**on by default**) lets the agent reply with a tappable single/multi-choice form
instead of a paragraph of questions â the natural move when it needs two or three facts before
it can answer. `compute_chart`, on by default too, is what lets it draw a chart; it computes
shares / growth / running totals / projections in code from numbers already in the material,
so the model never does the arithmetic. Pass `ask_forms=False` for an agent that should only
ever answer in prose.
alias = the **public human-readable slug** â strongly recommended: it becomes the memorable URL
segment for this agent (`{public_base}/t/<tenant>/<alias>`), the link you hand to people. Use a
URL-safe lowercase-hyphen name (`booking`, `suchka-venue`); it is normalised automatically.
Collisions / invalid names are reported in alias_result.
Input schema
{'type': 'object', 'title': 'create_agentArguments', 'required': ['name'], 'properties': {'name': {'type': 'string', 'title': 'Name'}, 'soul': {'type': 'string', 'title': 'Soul', 'default': ''}, 'task': {'type': 'string', 'title': 'Task', 'default': ''}, 'alias': {'type': 'string', 'title': 'Alias', 'default': ''}, 'model': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Model', 'default': None}, 'tools': {'anyOf': [{'type': 'array', 'items': {'type': 'string'}}, {'type': 'null'}], 'title': 'Tools', 'default': None}, 'skills': {'anyOf': [{'type': 'array', 'items': {'type': 'string'}}, {'type': 'null'}], 'title': 'Skills', 'default': None}, 'ask_forms': {'type': 'boolean', 'title': 'Ask Forms', 'default': True}, 'published': {'type': 'boolean', 'title': 'Published', 'default': False}, 'knowledge_bases': {'anyOf': [{'type': 'array', 'items': {'type': 'string'}}, {'type': 'null'}], 'title': 'Knowledge Bases', 'default': None}}}
Output schema
{'type': 'object', 'title': 'create_agentOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
create_knowledge_base
Create a knowledge base.
**Write `description` â it is what routes questions to this base.** Before each answer the
platform reads every attached base's `description` and decides which ones this question needs;
a base whose description is vague or blank gets searched when it should not, or missed when it
should. One line, in the words a visitor would use, naming the subject **and the boundary**:
"Country-by-country medical device registration requirements â not company or pricing
information." It is also the line shown in the console list, but that is the lesser job.
**Write `instructions` at creation time â do not leave it blank.** It is injected next to this
KB's excerpts whenever they are retrieved. It does NOT affect recall (recall is vector search +
max_distance); it governs how the model USES what was retrieved. Derive it from the expected
usage, one line each:
- Scope: what it covers / does not, and what to do when out of scope
("Covers residential mortgages only; for car or personal loans, say so and hand off").
- Authority: where it ranks ("Current company policy; overrides industry norms").
- Usage rules specific to this content ("Any quoted rate must state its effective date").
The generic "answer from the excerpts; say when not covered" is built into the platform â never
repeat it here. Blank is acceptable only for generic reference material with no special rules.
Worked examples: search_agent4_docs("knowledge base instructions examples").
max_distance is the relevance cutoff (blank = global default 0.6); tighten for regulatory
content, and calibrate against real queries rather than guessing.
top_k is how many passages this base contributes per turn (1-10, blank = 3). Set it from the
**shape of the documents**, not from how important the base feels: if a typical document is
five passages long, three of them is a third of a record, and the model answers from a
fragment. Measured on an 8,500-record regulation base whose median document is 5 passages â
at k=3 the right record was retrieved 36/40 times but only 4 answers had enough to go on; at
k=6, same hit rate, 12. Going to 10 added 3 more and 77% more material. More is not free:
material in the prompt suppresses tool calls, dose-dependently.
Input schema
{'type': 'object', 'title': 'create_knowledge_baseArguments', 'required': ['name'], 'properties': {'name': {'type': 'string', 'title': 'Name'}, 'top_k': {'anyOf': [{'type': 'integer'}, {'type': 'null'}], 'title': 'Top K', 'default': None}, 'description': {'type': 'string', 'title': 'Description', 'default': ''}, 'instructions': {'type': 'string', 'title': 'Instructions', 'default': ''}, 'max_distance': {'anyOf': [{'type': 'number'}, {'type': 'null'}], 'title': 'Max Distance', 'default': None}}}
Output schema
{'type': 'object', 'title': 'create_knowledge_baseOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
create_share
Create a **share / integration entry point** for an agent â this is how end users actually reach it.
**`published=True` only means "visible", not "reachable"**: for end users to talk to the agent
you must create a share. The response carries a directly openable chat link
(`{public_base}/s/<token>`) and the website embed URL (`{public_base}/embed/<token>`).
For a website widget, paste one line before </body>:
`<script src="{public_base}/embed.js" data-token="<token>"></script>`.
label names this entry point ("website widget", "support link"). Telegram/WhatsApp and other
channels are connected separately on the agent's Integration page in the console.
**No website?** Hand the returned `chat_url` or `qr_url` (QR code) straight to the tenant:
print it on business cards / flyers / in-store; scanning opens a full-page chat, no login,
returning visitors are remembered per browser.
**For links you give to humans, prefer `pretty_url`** (when present in the response):
`{public_base}/t/<tenant alias>/<agent alias>` â memorable, printable, survives token rotation.
No pretty_url = aliases not fully set â **fix that proactively**: agent alias via
`create_agent`'s alias param or `PUT /agents/{name}/alias`; tenant alias in console â Settings.
The `/s/<token>` link still works, but it is the machine/embed form, not one to read out to a
person.
Input schema
{'type': 'object', 'title': 'create_shareArguments', 'required': ['agent_name'], 'properties': {'label': {'type': 'string', 'title': 'Label', 'default': ''}, 'agent_name': {'type': 'string', 'title': 'Agent Name'}}}
Output schema
{'type': 'object', 'title': 'create_shareOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
create_skill
Create a skill. description = "when to use" (goes into the system prompt, keep it short);
instructions = the detailed guide (fetched on demand via load_skill).
The paradigm (important): instructions are visible only AFTER the model calls load_skill â if
calling some tool is **mandatory** ("user gives a phone number â must call save_contact"), the
trigger must be written into description, otherwise models often answer directly without
loading the skill and the rule never takes effect. Also never promise return values a tool
doesn't produce (ticket numbers / IDs) â the model will fabricate them. The returned warnings
field flags both patterns.
Input schema
{'type': 'object', 'title': 'create_skillArguments', 'required': ['name'], 'properties': {'name': {'type': 'string', 'title': 'Name'}, 'tools': {'anyOf': [{'type': 'array', 'items': {'type': 'string'}}, {'type': 'null'}], 'title': 'Tools', 'default': None}, 'description': {'type': 'string', 'title': 'Description', 'default': ''}, 'instructions': {'type': 'string', 'title': 'Instructions', 'default': ''}}}
Output schema
{'type': 'object', 'title': 'create_skillOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
create_storyline
Create a storyline draft. After creating, self-check with validate_storyline, then
publish_storyline.
user_visibility â what the end user sees of their own run: "invisible" (default, no UI),
"named" (a banner with the storyline name only), "trail" (banner + read-only view where
untaken branches and future steps are redacted grey blocks), "full" (banner with step x/y +
full read-only map). `learner_visibility` is the deprecated old name (legacy values
hidden/completed_only still accepted and mapped).
concurrency â who the progress follows: "user" (default) = progress belongs to the person, all
of that user's sessions share one run â fits curricula / onboarding / KYC; "session" =
progress belongs to the case, each session gets its own run, a new conversation = a new
application â fits licence applications / tickets / per-product flows. Case state goes to the
blackboard (travels with the run); facts about the person go to profile dimensions (shared
across runs).
graph = {"nodes":[Node,...], "edges":[]} (edges are derived from exits, may be left empty).
Node = {
node_key: stable uuid (unchanged across edits; exits/funnels reference it), title,
task (may interpolate {dimension}/{blackboard.key}),
type: "task"(default) | "document_review"(visual pre-review of uploads) |
"export"(structured report) | "parallel"(parallel branches / AND-join),
review: {"checkpoints":[str]} â type=document_review: per-item visual checkpoints,
export: {"sections":[str]} â type=export: report sections,
parallel: {"branches":[{"key":str,"label":str,"to_node_key":str},...]}
â type=parallel: declares required branches, each pointing at a sub-flow entry; the user
may do them in any order, the engine tracks completion, and only when ALL are done does
the node take its single join exit (put it at exits[0]). A branch sub-flow's last step
just exits back to this parallel node â no hand-written completion flags.
flags: {"is_entry":bool, "is_terminal":bool},
on_enter_opening: something to say proactively on entry (empty = silent transition),
ai_eval_trigger: natural-language condition for when to run AI evaluation (empty = every turn),
callback: {"mode":"none"|"backend"|"ui_redirect", "wait_timeout_secs":int, "signal_name":str},
profile_writes: [{"dim":str, "source":"ai"|"rule"|"callback"}] â dimensions this node writes,
resources: {"skills":[str],"knowledge_bases":[str],"tools":[str],
"resource_mode":"additive"|"replace"},
exits: [Exit,...] (list order = priority; deterministic rule/callback/user_choice are
evaluated first, ai last)
}
Exit = {"kind":..., "label":str, "to_node_key":str,
"ai_criteria":str â kind=ai: one natural-language criterion,
"user_choice":{"button_text":str} â kind=user_choice,
"rule_ast":RuleAst â kind=rule (see below; an AST, not a string),
"callback_signal":"done"|"timeout"|"canceled" â kind=callback,
"target_storyline_id":str, â kind=goto_storyline
"writes":[{"ref":"dim"|"blackboard","key":str,"op":"set"|"inc","value":<num|str|bool>},...]}
â deterministic state writes when this exit is taken (this is how profile_writes with
source='rule' actually land): set = assign (completion flags / branch flags),
inc = increment (loop/retry counters, value defaults to 1). dim writes are constrained
by profile_schema.
kind â ai|user_choice|rule|callback|goto_storyline.
Common control flow composes deterministically (never bet on the LLM): if/else/switch = several
rule exits on one node (order = priority); loop/retry = a back-edge + an inc counter in writes
+ a rule cap gate; AND-join = a parallel node, or hub + completion flags + an "and" rule.
RuleAst is one of:
comparison {"op":">="|">"|"<="|"<"|"=="|"!=",
"left":{"ref":"dim"|"blackboard","key":str}, "right":{"value":<num|str|bool>}}
boolean {"op":"and"|"or", "clauses":[RuleAst,...]}
profile_schema = dimension definitions, e.g.
{"listening":{"type":"int","min":0,"max":100,"visible_to_user":true}}.
on_complete="goto_next" requires next_storyline_id (validate/publish reject otherwise).
Enrolment (who enters this line, when): is_default=true auto-enrols on first conversation (at
most one per agent); allow_agent_enroll=true lets the agent enrol users mid-conversation â in
that case ALWAYS write enroll_trigger (one natural-language "enter when", e.g. "the visitor
says they want to apply for a loan"), otherwise the agent has no trigger basis and almost never
enrols; takes effect after publishing, independent of is_default and manual assignment.
See /docs/tenant-guide/storylines.
Input schema
{'type': 'object', 'title': 'create_storylineArguments', 'required': ['agent_name', 'key'], 'properties': {'key': {'type': 'string', 'title': 'Key'}, 'name': {'type': 'string', 'title': 'Name', 'default': ''}, 'entry': {'type': 'string', 'title': 'Entry', 'default': 'auto'}, 'graph': {'anyOf': [{'type': 'object', 'additionalProperties': True}, {'type': 'null'}], 'title': 'Graph', 'default': None}, 'agent_name': {'type': 'string', 'title': 'Agent Name'}, 'allow_exit': {'type': 'boolean', 'title': 'Allow Exit', 'default': False}, 'is_default': {'type': 'boolean', 'title': 'Is Default', 'default': False}, 'concurrency': {'type': 'string', 'title': 'Concurrency', 'default': 'user'}, 'description': {'type': 'string', 'title': 'Description', 'default': ''}, 'on_complete': {'type': 'string', 'title': 'On Complete', 'default': 'end'}, 'display_name': {'type': 'string', 'title': 'Display Name', 'default': ''}, 'show_profile': {'type': 'boolean', 'title': 'Show Profile', 'default': False}, 'enroll_trigger': {'type': 'string', 'title': 'Enroll Trigger', 'default': ''}, 'profile_schema': {'anyOf': [{'type': 'object', 'additionalProperties': True}, {'type': 'null'}], 'title': 'Profile Schema', 'default': None}, 'user_visibility': {'type': 'string', 'title': 'User Visibility', 'default': 'invisible'}, 'next_storyline_id': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Next Storyline Id', 'default': None}, 'allow_agent_enroll': {'type': 'boolean', 'title': 'Allow Agent Enroll', 'default': False}, 'learner_visibility': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Learner Visibility', 'default': None}}}
Output schema
{'type': 'object', 'title': 'create_storylineOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
diagnose_prompt
Read a field and return a list of **specific** improvement suggestions (string array).
field: same values as improve_prompt. current = the text to diagnose (required â an empty text
has nothing to diagnose). Typical flow: diagnose_prompt first, pick the suggestions you want,
then fold them into improve_prompt's hint.
Input schema
{'type': 'object', 'title': 'diagnose_promptArguments', 'required': ['field', 'current'], 'properties': {'field': {'type': 'string', 'title': 'Field'}, 'context': {'anyOf': [{'type': 'object', 'additionalProperties': True}, {'type': 'null'}], 'title': 'Context', 'default': None}, 'current': {'type': 'string', 'title': 'Current'}}}
Output schema
{'type': 'object', 'title': 'diagnose_promptOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
export_storyline
Export a portable storyline JSON (schema_version + definition; no tenant/version/id) for
backup, migration or re-import.
Input schema
{'type': 'object', 'title': 'export_storylineArguments', 'required': ['storyline_id'], 'properties': {'storyline_id': {'type': 'string', 'title': 'Storyline Id'}}}
Output schema
{'type': 'object', 'title': 'export_storylineOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
get_agent
Get one agent's full configuration (soul/task/tools/skills/knowledge_bases/model/params).
Input schema
{'type': 'object', 'title': 'get_agentArguments', 'required': ['name'], 'properties': {'name': {'type': 'string', 'title': 'Name'}}}
Output schema
{'type': 'object', 'title': 'get_agentOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
get_end_user
Get one end user's detail: basic profile (name, contact email/phone, city, timezone) + their
spaces.
Never any conversation content. Carries a `console_url` to the user's detail page.
Input schema
{'type': 'object', 'title': 'get_end_userArguments', 'required': ['uid'], 'properties': {'uid': {'type': 'string', 'title': 'Uid'}}}
Output schema
{'type': 'object', 'title': 'get_end_userOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
get_knowledge_base
Get one knowledge base's configuration (including the full instructions text shown to the model).
Input schema
{'type': 'object', 'title': 'get_knowledge_baseArguments', 'required': ['name'], 'properties': {'name': {'type': 'string', 'title': 'Name'}}}
Output schema
{'type': 'object', 'title': 'get_knowledge_baseOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
get_knowledge_index
Show the structured index profile: per column coverage, type, numeric range, top values, warnings.
**The warnings are the point.** On a real 4,500-document catalogue this surfaced three data
problems nobody knew about: 1,058 documents with `page_count` 0, 505 with `reading_level` 0,
493 with `word_count` 0 â those are not zeros, they are missing values recorded as 0, and they
silently poison every average, minimum and range filter built on them.
Report the warnings to the user in their own terms ("493 of your books have no word count â
they will all tie for shortest"). Nothing else in the platform will ever tell them.
Input schema
{'type': 'object', 'title': 'get_knowledge_indexArguments', 'required': ['name'], 'properties': {'name': {'type': 'string', 'title': 'Name'}}}
Output schema
{'type': 'object', 'title': 'get_knowledge_indexOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
get_skill
Get one skill's full content (including the complete instructions text).
Input schema
{'type': 'object', 'title': 'get_skillArguments', 'required': ['name'], 'properties': {'name': {'type': 'string', 'title': 'Name'}}}
Output schema
{'type': 'object', 'title': 'get_skillOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
get_storyline
Get one storyline's full definition (whole graph + profile_schema + top-level fields).
Input schema
{'type': 'object', 'title': 'get_storylineArguments', 'required': ['storyline_id'], 'properties': {'storyline_id': {'type': 'string', 'title': 'Storyline Id'}}}
Output schema
{'type': 'object', 'title': 'get_storylineOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
import_storyline
Import an exported storyline JSON as a draft. **Artifacts like "generate a storyline from a
novel / source material" land here.**
payload = the structure returned by export_storyline (schema_version + graph + profile_schema
+ top-level fields). Version-aware: payloads newer than the system supports are rejected.
Always creates new, never claims default, key collisions auto-rename.
override_agent_name rebinds the imported line to an agent that exists in the target tenant.
Input schema
{'type': 'object', 'title': 'import_storylineArguments', 'required': ['payload'], 'properties': {'payload': {'type': 'object', 'title': 'Payload', 'additionalProperties': True}, 'override_agent_name': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Override Agent Name', 'default': None}}}
Output schema
{'type': 'object', 'title': 'import_storylineOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
improve_prompt
Rewrite a field's text to work better as an AI system prompt; returns the improved text.
field â persona | task | greeting | storyline_task | storyline_opening
| storyline_ai_trigger | storyline_ai_criteria
| skill_description | skill_instructions | kb_description | kb_instructions
â determines the rewrite target (a Task should read like commands, a KB description says
"what's inside / when it's relevant", a skill description says "when to use" in one line, â¦).
current = the current text (empty = draft from scratch); hint = the direction you want this
time (empty = general polish); context = on-site context (which KB is being edited, values of
sibling fields) â the more specific, the better the fit.
Returns only the improved text, no explanation, no wrapper â the caller places it next to the
original for a human to accept or reject.
Input schema
{'type': 'object', 'title': 'improve_promptArguments', 'required': ['field'], 'properties': {'hint': {'type': 'string', 'title': 'Hint', 'default': ''}, 'field': {'type': 'string', 'title': 'Field'}, 'context': {'anyOf': [{'type': 'object', 'additionalProperties': True}, {'type': 'null'}], 'title': 'Context', 'default': None}, 'current': {'type': 'string', 'title': 'Current', 'default': ''}}}
Output schema
{'type': 'object', 'title': 'improve_promptOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
list_agents
List all agents in this tenant (name + published state).
Input schema
{'type': 'object', 'title': 'list_agentsArguments', 'properties': {}}
Output schema
{'type': 'object', 'title': 'list_agentsOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
list_end_users
List this tenant's **end users** (the people chatting with your agents) â roster only,
never any conversation content.
Each user carries: display name, login method (email / oauth:* / proxy), email (if any),
space/session/document counts, last-24h token usage, and a `console_url` (click through for
detail). q filters by name or external id; sort â recent (default) | tokens24h (descending
24h usage). Pagination: pass the returned next_cursor back as cursor.
Typical: list_end_users(sort="tokens24h") to find the most active / most expensive users.
Input schema
{'type': 'object', 'title': 'list_end_usersArguments', 'properties': {'q': {'type': 'string', 'title': 'Q', 'default': ''}, 'sort': {'type': 'string', 'title': 'Sort', 'default': 'recent'}, 'limit': {'type': 'integer', 'title': 'Limit', 'default': 50}, 'cursor': {'type': 'string', 'title': 'Cursor', 'default': ''}}}
Output schema
{'type': 'object', 'title': 'list_end_usersOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
list_knowledge_bases
List this tenant's knowledge bases (with doc/chunk counts and relevance cutoff).
Input schema
{'type': 'object', 'title': 'list_knowledge_basesArguments', 'properties': {}}
Output schema
{'type': 'object', 'title': 'list_knowledge_basesOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
list_mcp_servers
List the MCP servers connected in this tenant's tool library (read-only; secret fields omitted).
Input schema
{'type': 'object', 'title': 'list_mcp_serversArguments', 'properties': {}}
Output schema
{'type': 'object', 'title': 'list_mcp_serversOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
list_page_contexts
List all page playbooks of this tenant (with match rules, greeting mode, position).
Input schema
{'type': 'object', 'title': 'list_page_contextsArguments', 'properties': {}}
Output schema
{'type': 'object', 'title': 'list_page_contextsOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
list_shares
List an agent's shares (token / label / disabled / chat link / embed URL / appearance config).
When the top-level `pretty_url` is non-empty it is the preferred link to give humans
(see create_share).
Input schema
{'type': 'object', 'title': 'list_sharesArguments', 'required': ['agent_name'], 'properties': {'agent_name': {'type': 'string', 'title': 'Agent Name'}}}
Output schema
{'type': 'object', 'title': 'list_sharesOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
list_skills
List this tenant's skills.
Input schema
{'type': 'object', 'title': 'list_skillsArguments', 'properties': {}}
Output schema
{'type': 'object', 'title': 'list_skillsOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
list_storylines
List this tenant's storylines (optionally filtered by agent). Includes state, version, default flag.
Input schema
{'type': 'object', 'title': 'list_storylinesArguments', 'properties': {'agent': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Agent', 'default': None}}}
Output schema
{'type': 'object', 'title': 'list_storylinesOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
list_tools
List tool names assignable to agents, with their descriptions (incl. the tenant's connected MCP tools).
Input schema
{'type': 'object', 'title': 'list_toolsArguments', 'properties': {}}
Output schema
{'type': 'object', 'title': 'list_toolsOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
list_user_sessions
List one end user's **sessions** â metadata only: agent, title (AI summary), message count,
tokens, timestamps.
**Message bodies are not returned.** Transcripts are rendered in the console: each session
carries a `console_url` that **deep-links straight to that conversation, auto-opened** (no
digging through the detail page when there are many), plus `user_console_url` to the user's
detail page as the general entry. Hand the link to the tenant â a human reads the original
there. Pagination: pass next_cursor back as cursor.
Input schema
{'type': 'object', 'title': 'list_user_sessionsArguments', 'required': ['uid'], 'properties': {'uid': {'type': 'string', 'title': 'Uid'}, 'limit': {'type': 'integer', 'title': 'Limit', 'default': 50}, 'cursor': {'type': 'string', 'title': 'Cursor', 'default': ''}}}
Output schema
{'type': 'object', 'title': 'list_user_sessionsOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
page_context_stats
Open counts and suggested-question click counts per playbook â find copy nobody clicks.
Input schema
{'type': 'object', 'title': 'page_context_statsArguments', 'properties': {}}
Output schema
{'type': 'object', 'title': 'page_context_statsOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
patch_knowledge_index
Change the structured index schema from one sentence of plain language.
Example requests: "also track the author so users can find other books by them", "I want to
filter by illustrator", "drop the cover link". The model may only emit add/modify/remove/refused
â it cannot rewrite the table, so it has no way to damage columns the user did not mention.
**Data that is not in the documents comes back in `refused`, not as a new column.** Asked for
a publication year that the corpus does not carry, it says so instead of inventing an empty
column. Pass that refusal on to the user verbatim.
Call with `apply=false` first and show the user what would change; batch several edits and apply
once. Applying marks the index for rebuild â it does **not** re-embed anything, so it is cheap.
Input schema
{'type': 'object', 'title': 'patch_knowledge_indexArguments', 'required': ['name', 'request'], 'properties': {'name': {'type': 'string', 'title': 'Name'}, 'apply': {'type': 'boolean', 'title': 'Apply', 'default': False}, 'request': {'type': 'string', 'title': 'Request'}}}
Output schema
{'type': 'object', 'title': 'patch_knowledge_indexOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
preview_plan
Generate a plan for a goal and return it for review **without persisting anything** â no
temporary storyline, no enrollment, no user affected. Use it to inspect plan quality before
trusting the feature in production, or to see how the agent's `task` constrains the steps.
Returns {plan, backend, warnings}. `warnings` flags machine-detectable smells, notably checklist
items that are the AGENT's own deliverable ("produce a summary") â those can never be ticked off,
because ticking is judged from what the USER provides, so the step would stall forever.
Input schema
{'type': 'object', 'title': 'preview_planArguments', 'required': ['agent', 'goal'], 'properties': {'goal': {'type': 'string', 'title': 'Goal'}, 'agent': {'type': 'string', 'title': 'Agent'}}}
Output schema
{'type': 'object', 'title': 'preview_planOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
publish_storyline
Publish a storyline: validates first (blocking errors â 422), then freezes an immutable
version and marks it published.
Input schema
{'type': 'object', 'title': 'publish_storylineArguments', 'required': ['storyline_id'], 'properties': {'storyline_id': {'type': 'string', 'title': 'Storyline Id'}}}
Output schema
{'type': 'object', 'title': 'publish_storylineOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
resolve_page_context
Given a URL (or key), show which playbook it resolves to â always verify after writing a
match rule.
Globs make it easy to write rules that "look right but never match" (one `*` short, one path
level too many) â and in production the only symptom is visitors silently getting the default
playbook, with no error anywhere.
Input schema
{'type': 'object', 'title': 'resolve_page_contextArguments', 'properties': {'key': {'type': 'string', 'title': 'Key', 'default': ''}, 'url': {'type': 'string', 'title': 'Url', 'default': ''}}}
Output schema
{'type': 'object', 'title': 'resolve_page_contextOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
search_agent4_docs
Search agent4.io's own product docs â concepts (what a thing *is*) and cookbook recipes (which
tool to *call*). Use when you're unsure how an agent4.io feature works, what a term means, or how to
build something on the platform. Returns top passages with title, layer ("concept"|"ops"), url and a
snippet. Public content; no tenant data involved.
Input schema
{'type': 'object', 'title': 'search_agent4_docsArguments', 'required': ['query'], 'properties': {'k': {'type': 'integer', 'title': 'K', 'default': 5}, 'query': {'type': 'string', 'title': 'Query'}}}
Output schema
{'type': 'object', 'title': 'search_agent4_docsOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
search_knowledge_base
Search this knowledge base; returns matching chunks with distances.
Uses **exactly the same retrieval path as real conversations** â use it to validate the cutoff
and chunking: an empty result means this question will be judged "not covered" and the agent
will answer "not in the knowledge base".
Input schema
{'type': 'object', 'title': 'search_knowledge_baseArguments', 'required': ['kb_name', 'query'], 'properties': {'query': {'type': 'string', 'title': 'Query'}, 'kb_name': {'type': 'string', 'title': 'Kb Name'}}}
Output schema
{'type': 'object', 'title': 'search_knowledge_baseOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
set_custom_domain
Bind the client's own domain to this tenant's chat pages: `https://chat.client.com/` serves
the tenant's branded page (or a specific agent's chat), the address bar keeps the client's
domain, and the TLS certificate is issued automatically.
**Have the client configure DNS first**: add a CNAME at their DNS provider pointing to the
`cname_target` in the response (`endpoint.agent4.io`). Key points:
- **Subdomains only** (`chat.client.com`). An apex domain (`client.com`) cannot carry a CNAME â
have the client use a subdomain, or a DNS provider with CNAME flattening (Cloudflare etc.).
- **Cloudflare users must set the record to DNS only (grey cloud)**. With the proxy on (orange
cloud) the name resolves to Cloudflare's addresses, verification fails, and `last_error`
says so.
- Verification is **asynchronous**: checked once immediately on binding; if DNS hasn't
propagated it stays `pending` and is re-checked every 10 minutes (calling this tool again
also re-triggers). `status` = `active` means live; the certificate is issued on first visit.
- **One** domain per tenant; changing the domain re-runs verification, changing only
`agent_alias` does not.
- Requires a plan that includes custom domains (403 = upgrade needed).
`agent_alias` empty â the domain lands on the tenant's branded page (listing available
agents); set to an agent's public alias â lands directly on that agent's chat page.
Input schema
{'type': 'object', 'title': 'set_custom_domainArguments', 'required': ['domain'], 'properties': {'domain': {'type': 'string', 'title': 'Domain'}, 'agent_alias': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Agent Alias', 'default': None}}}
Output schema
{'type': 'object', 'title': 'set_custom_domainOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
set_pwa_branding
Configure a specific **agent's** PWA branding and install experience (applies to that
agent's `/s/` standalone chat pages â what gets installed to the home screen is one agent's
entry page, so each agent is its own app).
- `agent`: the agent name whose install branding to configure (required).
- `icon_source_url`: public URL of one master image (PNG/JPG/WebP â¥192Ã192, ideally a square
logo). The server derives the full set: browser-tab favicon(48) + install icons 192/512 +
Android maskable(512). Non-square images are centre-cropped. Once set, "Add to Home Screen"
installs this agent's own icon.
- `install_prompt`: prompt style â `banner` (a dismissible slim bar inside the chat page,
default), `card` (a card shown on first visit; more visible, more intrusive), `off` (no
prompt). Android/Chrome uses the system install dialog; iOS automatically switches to a
"Share â Add to Home Screen" illustrated guide.
Icon/prompt params optional, settable independently. Returns the final config (icon URLs +
install_prompt). Full walkthrough in the Cookbook's configure-branding.
Input schema
{'type': 'object', 'title': 'set_pwa_brandingArguments', 'required': ['agent'], 'properties': {'agent': {'type': 'string', 'title': 'Agent'}, 'install_prompt': {'type': 'string', 'title': 'Install Prompt', 'default': ''}, 'icon_source_url': {'type': 'string', 'title': 'Icon Source Url', 'default': ''}}}
Output schema
{'type': 'object', 'title': 'set_pwa_brandingOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
tenant_info
This tenant's basic info and quota (plus custom-domain binding status, if any).
Input schema
{'type': 'object', 'title': 'tenant_infoArguments', 'properties': {}}
Output schema
{'type': 'object', 'title': 'tenant_infoOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
test_planner_trigger
Dry-run the Dynamic Planner's escalation judge: would these messages be offered a
step-by-step plan? Use it after editing the agent's `task`, since the task text is the
**business-domain gate** â a vague task lets out-of-domain requests through, a narrow one
turns real customers away.
messages: the user's turns in order; the LAST one is treated as the current turn and the rest
as that user's earlier messages (same window the live judge sees). samples>1 re-runs the judge
to show stability (it is a probabilistic call). expect: pass true/false to get `matched`.
Nothing is stored â no session, no enrollment. Tokens count toward the tenant quota (kind=planner).
Returns {triggered, samples, goals[], domain, matched}. Reads the judge only: whether the user
then ACCEPTS the offer is a separate, user-controlled step.
Input schema
{'type': 'object', 'title': 'test_planner_triggerArguments', 'required': ['agent', 'messages'], 'properties': {'agent': {'type': 'string', 'title': 'Agent'}, 'expect': {'anyOf': [{'type': 'boolean'}, {'type': 'null'}], 'title': 'Expect', 'default': None}, 'samples': {'type': 'integer', 'title': 'Samples', 'default': 1}, 'messages': {'type': 'array', 'items': {'type': 'string'}, 'title': 'Messages'}}}
Output schema
{'type': 'object', 'title': 'test_planner_triggerOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
test_skill_trigger
Reality-check whether your prompts actually trigger tool calls (dry-run) â run this after
writing/changing a skill instead of counting corpses in production.
Replays your messages N times against the **production** system-prompt assembly, tool schemas
and this tenant's actual model routing, capturing only the model's tool-call decision: **tool
side effects are NOT executed**, no session is stored. Tokens count toward the tenant quota
(messagesâ¤5, samplesâ¤5, at most 25 calls per invocation â pick test messages carefully).
Two modes for the skill's two battlefields:
- loaded=false (default): first turn, skill not loaded â tests whether the trigger in
description works;
- loaded=true: simulates post-load_skill â tests the quality of instructions (incl. few-shot
examples).
Returns per-message hit counts plus claimed_without_call (the model said "noted" WITHOUT
calling the tool â the worst failure, fix first). Cover edge cases in your test messages:
numbers with spaces, buried in long questions, corrections, email-only. The loop:
create_skill â check warnings (static lint) â test_skill_trigger (dynamic reality check) â
adjust description / add examples â re-test until the hit rate holds.
Input schema
{'type': 'object', 'title': 'test_skill_triggerArguments', 'required': ['agent', 'messages'], 'properties': {'agent': {'type': 'string', 'title': 'Agent'}, 'skill': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Skill', 'default': None}, 'loaded': {'type': 'boolean', 'title': 'Loaded', 'default': False}, 'samples': {'type': 'integer', 'title': 'Samples', 'default': 3}, 'messages': {'type': 'array', 'items': {'type': 'string'}, 'title': 'Messages'}, 'expect_tool': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Expect Tool', 'default': None}}}
Output schema
{'type': 'object', 'title': 'test_skill_triggerOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
unpublish_storyline
Unpublish (archive): no more auto-enrolment; in-flight users silently degrade to the plain
agent on their next turn (progress kept â re-publishing resumes it).
Input schema
{'type': 'object', 'title': 'unpublish_storylineArguments', 'required': ['storyline_id'], 'properties': {'storyline_id': {'type': 'string', 'title': 'Storyline Id'}}}
Output schema
{'type': 'object', 'title': 'unpublish_storylineOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
update_agent
Update **part** of an agent â fields you don't pass stay as they are.
â ï¸ List-field semantics, don't mix them up: `tools=[...]` / `skills=[...]` /
`knowledge_bases=[...]` are **full-list replacement** â items you didn't read first get pushed
out. To "add one / remove one", use `add_tools` / `remove_tools` (and add_/remove_skills,
add_/remove_knowledge_bases): incremental, idempotent, leaves the rest untouched.
After the call, **read the response** to verify the final list â don't announce "enabled" just
because the call succeeded.
Also: tools can ride on a skill (the tools field of create_skill/update_skill) â once the skill
is attached to an agent, its bound tools join the chat-time whitelist automatically, but they
**do not appear in the agent's own tools list**.
`auto_suggest=True` makes the agent propose 1â3 clickable follow-up questions after every turn
(rendered above the composer; clicking one sends it). `suggest_prompt` is optional steering for
those questions ("always move toward booking a demo") â leave it empty and they are written
from the conversation alone. Costs one extra short LLM call per turn.
Input schema
{'type': 'object', 'title': 'update_agentArguments', 'required': ['name'], 'properties': {'name': {'type': 'string', 'title': 'Name'}, 'soul': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Soul', 'default': None}, 'task': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Task', 'default': None}, 'model': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Model', 'default': None}, 'tools': {'anyOf': [{'type': 'array', 'items': {'type': 'string'}}, {'type': 'null'}], 'title': 'Tools', 'default': None}, 'skills': {'anyOf': [{'type': 'array', 'items': {'type': 'string'}}, {'type': 'null'}], 'title': 'Skills', 'default': None}, 'greeting': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Greeting', 'default': None}, 'add_tools': {'anyOf': [{'type': 'array', 'items': {'type': 'string'}}, {'type': 'null'}], 'title': 'Add Tools', 'default': None}, 'ask_forms': {'anyOf': [{'type': 'boolean'}, {'type': 'null'}], 'title': 'Ask Forms', 'default': None}, 'published': {'anyOf': [{'type': 'boolean'}, {'type': 'null'}], 'title': 'Published', 'default': None}, 'reasoning': {'anyOf': [{'type': 'boolean'}, {'type': 'null'}], 'title': 'Reasoning', 'default': None}, 'add_skills': {'anyOf': [{'type': 'array', 'items': {'type': 'string'}}, {'type': 'null'}], 'title': 'Add Skills', 'default': None}, 'auto_suggest': {'anyOf': [{'type': 'boolean'}, {'type': 'null'}], 'title': 'Auto Suggest', 'default': None}, 'remove_tools': {'anyOf': [{'type': 'array', 'items': {'type': 'string'}}, {'type': 'null'}], 'title': 'Remove Tools', 'default': None}, 'auto_greeting': {'anyOf': [{'type': 'boolean'}, {'type': 'null'}], 'title': 'Auto Greeting', 'default': None}, 'remove_skills': {'anyOf': [{'type': 'array', 'items': {'type': 'string'}}, {'type': 'null'}], 'title': 'Remove Skills', 'default': None}, 'suggest_prompt': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Suggest Prompt', 'default': None}, 'knowledge_bases': {'anyOf': [{'type': 'array', 'items': {'type': 'string'}}, {'type': 'null'}], 'title': 'Knowledge Bases', 'default': None}, 'grounding_required': {'anyOf': [{'type': 'boolean'}, {'type': 'null'}], 'title': 'Grounding Required', 'default': None}, 'add_knowledge_bases': {'anyOf': [{'type': 'array', 'items': {'type': 'string'}}, {'type': 'null'}], 'title': 'Add Knowledge Bases', 'default': None}, 'remove_knowledge_bases': {'anyOf': [{'type': 'array', 'items': {'type': 'string'}}, {'type': 'null'}], 'title': 'Remove Knowledge Bases', 'default': None}}}
Output schema
{'type': 'object', 'title': 'update_agentOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
update_knowledge_base
Update part of a knowledge base â fields you don't pass stay as they are (server-side PATCH merge).
Input schema
{'type': 'object', 'title': 'update_knowledge_baseArguments', 'required': ['name'], 'properties': {'name': {'type': 'string', 'title': 'Name'}, 'top_k': {'anyOf': [{'type': 'integer'}, {'type': 'null'}], 'title': 'Top K', 'default': None}, 'description': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Description', 'default': None}, 'instructions': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Instructions', 'default': None}, 'max_distance': {'anyOf': [{'type': 'number'}, {'type': 'null'}], 'title': 'Max Distance', 'default': None}}}
Output schema
{'type': 'object', 'title': 'update_knowledge_baseOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
update_skill
Update part of a skill â fields you don't pass stay as they are (server-side PATCH merge,
no GET+PUT round-trip).
â ï¸ `tools=[...]` is **full-list replacement**; to add/remove one use `add_tools`/`remove_tools`
(incremental, idempotent). Tools bound to a skill take effect automatically at chat time on any
agent the skill is attached to â no need to also add them to the agent's tools.
Input schema
{'type': 'object', 'title': 'update_skillArguments', 'required': ['name'], 'properties': {'name': {'type': 'string', 'title': 'Name'}, 'tools': {'anyOf': [{'type': 'array', 'items': {'type': 'string'}}, {'type': 'null'}], 'title': 'Tools', 'default': None}, 'add_tools': {'anyOf': [{'type': 'array', 'items': {'type': 'string'}}, {'type': 'null'}], 'title': 'Add Tools', 'default': None}, 'description': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Description', 'default': None}, 'instructions': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Instructions', 'default': None}, 'remove_tools': {'anyOf': [{'type': 'array', 'items': {'type': 'string'}}, {'type': 'null'}], 'title': 'Remove Tools', 'default': None}}}
Output schema
{'type': 'object', 'title': 'update_skillOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
update_storyline
Update **part** of a storyline draft â fields you don't pass stay as they are (GET-then-merge
on top of a full PUT underneath).
concurrency: "user" = progress belongs to the person (shared across sessions); "session" =
progress belongs to the case (one run per session). Changes affect only **future** enrolments;
in-flight runs are not migrated.
â ï¸ Especially `graph`: not passing it = keep the existing graph. (This tool once treated "no
graph" as replace-with-empty â renaming a storyline wiped its whole flow. Semantics are now
partial; to truly clear the graph, pass `{"nodes":[],"edges":[]}` explicitly.)
graph/Node/Exit/RuleAst structures: see create_storyline. Pass expected_version for optimistic
locking. Read the response to verify, then validate_storyline.
Input schema
{'type': 'object', 'title': 'update_storylineArguments', 'required': ['storyline_id', 'agent_name', 'key'], 'properties': {'key': {'type': 'string', 'title': 'Key'}, 'name': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Name', 'default': None}, 'entry': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Entry', 'default': None}, 'graph': {'anyOf': [{'type': 'object', 'additionalProperties': True}, {'type': 'null'}], 'title': 'Graph', 'default': None}, 'agent_name': {'type': 'string', 'title': 'Agent Name'}, 'allow_exit': {'anyOf': [{'type': 'boolean'}, {'type': 'null'}], 'title': 'Allow Exit', 'default': None}, 'is_default': {'anyOf': [{'type': 'boolean'}, {'type': 'null'}], 'title': 'Is Default', 'default': None}, 'concurrency': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Concurrency', 'default': None}, 'description': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Description', 'default': None}, 'on_complete': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'On Complete', 'default': None}, 'display_name': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Display Name', 'default': None}, 'show_profile': {'anyOf': [{'type': 'boolean'}, {'type': 'null'}], 'title': 'Show Profile', 'default': None}, 'storyline_id': {'type': 'string', 'title': 'Storyline Id'}, 'enroll_trigger': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Enroll Trigger', 'default': None}, 'profile_schema': {'anyOf': [{'type': 'object', 'additionalProperties': True}, {'type': 'null'}], 'title': 'Profile Schema', 'default': None}, 'user_visibility': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'User Visibility', 'default': None}, 'expected_version': {'anyOf': [{'type': 'integer'}, {'type': 'null'}], 'title': 'Expected Version', 'default': None}, 'next_storyline_id': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Next Storyline Id', 'default': None}, 'allow_agent_enroll': {'anyOf': [{'type': 'boolean'}, {'type': 'null'}], 'title': 'Allow Agent Enroll', 'default': None}, 'learner_visibility': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Learner Visibility', 'default': None}}}
Output schema
{'type': 'object', 'title': 'update_storylineOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
upsert_page_context
Create or fully replace a page playbook.
`context` is the page background the **server** injects for the agent â write "who lands on
this page, what they are deciding, what they usually worry about"; do NOT write facts like
prices or quotas (those belong in a knowledge base, which also ranks higher in authority).
`url_pattern` is a glob (`*/pricing`, `*/solutions/*`), matches the path only, ignores query
strings and trailing slashes; without it the page must report the key explicitly. Resolution
order: explicit key > url_pattern > default.
`greeting_mode="generated"` produces the greeting and suggested questions on the fly in the
visitor's language (recommended); `"static"` uses your fixed `greeting` / `questions`.
**This is a full replace**: fields you don't pass fall back to defaults rather than staying
as they are. To change one field, `list_page_contexts` first, merge, then send.
Input schema
{'type': 'object', 'title': 'upsert_page_contextArguments', 'required': ['key'], 'properties': {'key': {'type': 'string', 'title': 'Key'}, 'label': {'type': 'string', 'title': 'Label', 'default': ''}, 'context': {'type': 'string', 'title': 'Context', 'default': ''}, 'greeting': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Greeting', 'default': None}, 'position': {'type': 'integer', 'title': 'Position', 'default': 100}, 'questions': {'anyOf': [{'type': 'array', 'items': {'type': 'string'}}, {'type': 'null'}], 'title': 'Questions', 'default': None}, 'is_default': {'type': 'boolean', 'title': 'Is Default', 'default': False}, 'url_pattern': {'anyOf': [{'type': 'string'}, {'type': 'null'}], 'title': 'Url Pattern', 'default': None}, 'greeting_mode': {'type': 'string', 'title': 'Greeting Mode', 'default': 'generated'}}}
Output schema
{'type': 'object', 'title': 'upsert_page_contextOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
usage_stats
Usage statistics. group â total | agent | user | space | day | model.
Metadata only â token counts and event counts, never any conversation content.
Use group="user" to see who uses the most, group="agent" for the busiest agent.
Input schema
{'type': 'object', 'title': 'usage_statsArguments', 'properties': {'group': {'type': 'string', 'title': 'Group', 'default': 'total'}}}
Output schema
{'type': 'object', 'title': 'usage_statsOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
validate_storyline
Validate the storyline graph (entry / dead ends / unreachable / dangling / cross-line targets
/ rule dimensions). Returns {ok, errors}. Must pass before publishing.
Input schema
{'type': 'object', 'title': 'validate_storylineArguments', 'required': ['storyline_id'], 'properties': {'storyline_id': {'type': 'string', 'title': 'Storyline Id'}}}
Output schema
{'type': 'object', 'title': 'validate_storylineOutput', 'required': ['result'], 'properties': {'result': {'type': 'string', 'title': 'Result'}}}
History
Recent tool changes
Added
import_storyline
Sept. 17, 2026, 12:39 p.m.
Added
export_storyline
Sept. 17, 2026, 12:39 p.m.
Added
clone_storyline
Sept. 17, 2026, 12:39 p.m.
Added
unpublish_storyline
Sept. 17, 2026, 12:39 p.m.
Added
publish_storyline
Sept. 17, 2026, 12:39 p.m.
Added
validate_storyline
Sept. 17, 2026, 12:39 p.m.
Added
update_storyline
Sept. 17, 2026, 12:39 p.m.
Added
create_storyline
Sept. 17, 2026, 12:39 p.m.
Added
get_storyline
Sept. 17, 2026, 12:39 p.m.
Added
list_storylines
Sept. 17, 2026, 12:39 p.m.
Added
page_context_stats
Sept. 17, 2026, 12:39 p.m.
Added
resolve_page_context
Sept. 17, 2026, 12:39 p.m.
Added
upsert_page_context
Sept. 17, 2026, 12:39 p.m.
Added
list_page_contexts
Sept. 17, 2026, 12:39 p.m.
Added
list_user_sessions
Sept. 17, 2026, 12:39 p.m.
Added
get_end_user
Sept. 17, 2026, 12:39 p.m.
Added
list_end_users
Sept. 17, 2026, 12:39 p.m.
Added
search_agent4_docs
Sept. 17, 2026, 12:39 p.m.
Added
tenant_info
Sept. 17, 2026, 12:39 p.m.
Added
usage_stats
Sept. 17, 2026, 12:39 p.m.
Added
preview_plan
Sept. 17, 2026, 12:39 p.m.
Added
test_planner_trigger
Sept. 17, 2026, 12:39 p.m.
Added
diagnose_prompt
Sept. 17, 2026, 12:39 p.m.
Added
improve_prompt
Sept. 17, 2026, 12:39 p.m.
Added
patch_knowledge_index
Sept. 17, 2026, 12:39 p.m.
Added
get_knowledge_index
Sept. 17, 2026, 12:39 p.m.
Added
build_knowledge_index
Sept. 17, 2026, 12:39 p.m.
Added
search_knowledge_base
Sept. 17, 2026, 12:39 p.m.
Added
add_knowledge_file
Sept. 17, 2026, 12:39 p.m.
Added
add_knowledge_text
Sept. 17, 2026, 12:39 p.m.
Discover
Similar MCP servers
GPT55 Model Gateway
xyz.558686.gpt55/token-gateway
Acts as a remote model gateway exposing GPT chat, coding and review services, lightweight text utilities, and prepaid or x402-bas…
VaultCrux Memory Core
io.github.CueCrux/vaultcrux-memory-core
Stores and evaluates agent memory, facts, decisions, constraints, expertise, checkpoints, contradictions, and delegation readines…
AgentBoard
app.agentsknow/agentboard
Provides agents with private memory, knowledge pages, evidence, posts, tasks, coordination spaces, proposals, voting, and resumab…
MeshMarket
io.github.RightOnPar-LLC/mesh-connector
Provides an agent-to-agent marketplace for reasoning, persistent memory, media generation, code review, audits, business tools, a…
MeshMarket
io.github.RightOnPar-LLC/meshmarket
Marketplace and gateway for metered agent capabilities including memory, reasoning, code review, research, media generation, busi…
DDG Agent Services
io.github.daedalusdevelopmentgroup/ddg-agent-services-mcp
Acts as a pay-per-call gateway for agent tools including LLM routes, web and market data, Ethereum RPC, security scans, service c…
x711 — Universal Agent Gas Station
io.github.0580iris-lang/x711-gas-station
Provides agents with web observation and interaction, search and retrieval, sandboxed computation, inter-agent communication, col…
Hipocampo MCP
io.github.carrasquelalex1/hipocampo
Provides bilingual semantic memory, embeddings, graph links, code indexing and search, context preload, deduplication, compressio…