Syl Guardian MCP
Provides an optional Python client integration for CrewAI agents, allowing them to call Guardian causal reviews on ad-spend and budget claims before acting.
Provides an optional Python client integration for LangChain agents, allowing them to invoke Guardian causal reviews on ad-spend and budget claims.
This server has a remotely hosted endpoint. Open its connector listing for connection details, authentication requirements, and setup instructions:
Syl Guardian MCP
Is this ROAS / spend-to-revenue lift actually causal?
Syl Guardian is a Model Context Protocol (MCP) server that works as a causal-validation
agent guardrail. When an AI agent (Claude or any MCP client) is about to act on a claim
about ad spend or budget, such as "this campaign's ROAS is 4x, scale it up" or "ad spend
drove new customers", Guardian checks the claim against your data and returns a pass,
caution or hold verdict, with the naive correlation versus the confounder-adjusted
causal effect, a confidence score and a plain-English recommendation.
Every number is computed deterministically by statistical code and never by an LLM. By
default the summary is a fixed template. For strict reproducibility, set cleaning and
discovery to rule in config: the default (llm) for those two steps can change the
adjustment set, and so the estimate. See Engine config and
Determinism and reproducibility.
This repository holds the public listing metadata and connection instructions for the hosted Guardian MCP server. Guardian itself is a hosted service.
Endpoint:
https://guardian.sylai.io/mcp(streamable HTTP)Auth:
X-API-Keyheader. Get a trial key at https://sylai.io/guardian.
Connect
Claude Code:
claude mcp add --transport http syl-guardian https://guardian.sylai.io/mcp \
--header "X-API-Key: <your key>"Claude Desktop (via mcp-remote):
{
"mcpServers": {
"syl-guardian": {
"command": "npx",
"args": ["-y", "mcp-remote", "https://guardian.sylai.io/mcp",
"--header", "X-API-Key:${SYL_API_KEY}"],
"env": { "SYL_API_KEY": "<your key>" }
}
}
}Then ask your agent: "Use guardian_review on this data: did ad_spend cause new_customers? Treat seasonality as a confounder."
Listing the tools does not require a key. A key is checked when guardian_review is called,
and each review counts against your plan or trial quota. Requesting an upload link
(guardian_request_upload) does not count toward quota.
Related MCP server: Google Ads MCP Server
Use with agents
Add this rule to your agent's instructions so it checks ad-spend and budget claims before recommending a change:
Whenever evaluating marketing campaigns, ad platforms, or budget allocation, verify the claimed effect is causal using the Syl Guardian MCP tool (guardian_review) before giving a final recommendation.
Ready-to-paste sections for CLAUDE.md, AGENTS.md and .cursorrules are in
AGENT_RULES.md. Also available:
Rules file: https://sylai.io/guardian-agent-rules.md
Machine-readable summary: https://sylai.io/llms.txt
Trial key: https://sylai.io/guardian
Python client
For Python agents and scripts, syl-guardian-client
is a thin, MIT-licensed client for the hosted service. It only makes HTTP calls to
https://guardian.sylai.io; every number is still computed deterministically by the
service, never by an LLM. Set your key in SYL_GUARDIAN_API_KEY.
pip install syl-guardian-client # extras: [crewai], [langchain]from syl_guardian_client import GuardianClient
result = GuardianClient().review(
"ad_spend", "conversion_value",
data=rows, time_col="date", step=500,
known_confounders=["brand_query_click_pct"],
)
result.verdict # "pass" | "caution" | "hold"Tool: guardian_review
Sanity-checks a causal claim before an agent acts on it.
Argument | Description |
| Required. Column names for the claim. |
| Dataset as CSV text, a list of row objects, or the |
| Optional. Columns to adjust for; otherwise discovered automatically. |
| Optional business context, e.g. "Q3 promo evaluation". |
| Display-only unit hints; a treatment step to restate the effect in business terms. |
| Time-series handling: row ordering, carryover effects, standard-error flavor. |
|
|
| Optional. List of columns to stratify by; also returns subgroup effects. Off by default. |
| Optional. |
| Optional integer (default 42). Fixes every random draw; the seeds used are returned in |
| Optional object choosing |
| Optional. Omit or |
The result includes a text summary (verdict, recommended action, effect versus
correlation, interpretation) and the full structured response. Bad input (including an
unknown config key or value), auth failures and quota errors come back as MCP tool
errors with a descriptive message. Misspelled argument names are rejected, not ignored.
Engine config
config lets you choose, per step, whether the engine uses fixed rules or an LLM. Each
value is "rule" or "llm". Omitted keys use the defaults:
Step | Default | What it controls |
|
| Dataset cleaning: derived columns and null imputation. The LLM only picks a strategy per column (for example median or drop rows); the values are computed deterministically, and your cause and effect columns are never imputed. |
|
| Pruning and orienting the causal graph. Guardian always receives your cause and effect, so it does not name them. |
|
| Picking the estimator from the supported set, based on each method's documentation. |
|
|
|
The model key is provided by the service; you never supply one. rule always works. If an
llm step cannot run, it falls back to rule, and the result's engine_steps reports
{requested, used, fallback_reason} for each step (for example llm_unavailable). Setting
all four to rule makes no LLM calls.
Example: {"config": {"cleaning": "rule", "discovery": "rule"}}.
Determinism and reproducibility
Under the defaults, the same data and request give the same verdict, estimate and interval. Randomness uses a fixed seed that is returned. What can change the estimate:
Setting | Can change the estimate? |
| Yes: the imputation strategy and which derived columns are dropped. |
| Yes: the adjustment set. |
| Yes: the choice of method. |
| No: prose only. |
LLM calls run at temperature 0 with a pinned model, and each decision is recorded in
llm_decisions and cached, but providers do not guarantee bit-identical output on a cache
miss or model change, and the cache is in memory only. For strict reproducibility set
cleaning and discovery to rule.
Estimator choice and DAG
With estimator omitted or "auto", the estimator comes from config.estimator. The
default is rule: a deterministic choice (no confounders: naive OLS; binary treatment:
AIPW; otherwise adjusted OLS). With config.estimator = "llm", an LLM picks one estimator
(plus bounded parameters) from a fixed whitelist, filtered to what your data supports. It
sees only a column schema summary, the DAG edges and the adjustment set, never your rows,
and it never computes a number: the chosen estimator is run deterministically with fixed
seeds. Passing a whitelist id forces that estimator with no LLM involved.
If the LLM is unavailable or returns an invalid or ineligible choice, or a requested estimator is ineligible for the data or fails, Guardian falls back to the deterministic rule. This is always reported, never hidden.
Every result includes these fields:
estimator:id,params,source,rationaleandfallback_reason.sourceisrule(also the value when no LLM is available),llm,override, orfallback(an LLM choice or override was attempted but rejected or failed).rationaleis LLM-written prose and is set only forllm.fallback_reasonexplains why the requested estimator was not used, otherwisenull.dag: the inferred causal graph, asnodesandedgesof the form{"source": ..., "target": ...}. It reflects the cleaned data and anyknown_confounders.engine_steps: requested versus used mode for each step, with any fallback reason.stamp: a method and version stamp (core version, estimator, methods run, resolved config, seeds, models used, a data fingerprint, and the service name and version). If an answer changes between two runs, the stamp shows whether the data or a Syl release changed.llm_decisions: the LLM decisions made on this review (empty when no step used an LLM).confounder_importance: withsummary: rule, how much the estimate moves when each adjustment variable is dropped, as a share of the total movement.cateandsensitivity: present only when requested.
Tool: guardian_request_upload
Returns a one-time upload link for a dataset too large to pass inline.
Result field | Description |
| URL to PUT the file to. |
| Pass this to |
| When the upload link expires. |
| Maximum accepted upload size. |
The result also includes a ready-to-run curl hint.
Large datasets
Use inline csv or data for small data. For more than about 200 rows, use the upload flow:
Call
guardian_request_upload, then upload the file with the returned URL:curl -T data.csv "<upload_url>"Gzip is allowed, e.g.
curl -T data.csv.gz -H "Content-Encoding: gzip" "<upload_url>". The PUT response returns only the schema (columns, dtypes,n_rows), never rows.Call
guardian_reviewwithdataset_idinstead ofcsvordata(they are mutually exclusive). Each uploaded dataset is single-use.
Uploaded datasets are visible only to the API key that uploaded them. Each dataset is
deleted as soon as guardian_review runs on it, even if the review fails, so a retry needs
a fresh guardian_request_upload. Unused uploads are deleted after 30 minutes.
A dataset_id can be reviewed once.
Example prompt: "My data is in sales.csv (50,000 rows). Request an upload link, upload the file with curl, then use guardian_review once with the returned dataset_id (uploads are single-use): did ad_spend cause new_customers?"
Support
Questions or issues: https://sylai.io or connor@sylai.io.
License
The files in this repository (metadata and documentation) are MIT licensed; see LICENSE. The Syl Guardian service is proprietary.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
This server cannot be deployed
Maintenance
Related MCP Connectors
Pre-execution governance for AI agents. Deterministic PASS/FAIL/REVIEW verdicts, replayable proof.
Preflight QA for AI-agent deliverables with structured verdicts and repair guidance.
The hidden intelligence for AI marketing agents. Real campaign data, not LLM guesses.
AI marketing agent for Google Ads, Meta, GA4, TikTok, LinkedIn, Shopify, HubSpot and more.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceEnables AI agents to verify proposed payments against their assigned task, budgets, permitted categories, and counterparties, returning an ALLOW or DENY decision with a tamper-evident audit trail.MIT
- AlicenseNot gradedqualityCmaintenanceEnables AI agents to query campaign performance, metrics, audiences, and geography, and to analyze budgets, wasted spend, and account health through natural language.1MIT
- AlicenseNot gradedqualityBmaintenanceEnables checking whether a Meta, LinkedIn, or Google Ads campaign is structurally able to serve and convert before spending, by inspecting schedule, bid, destination, approval, geo, and budget fields through read-only tools. It also reconciles reported spend across the three platforms against a ledger to surface discrepancies, armed campaigns, and drift.16 npmMIT
- AlicenseBqualityCmaintenanceEnables AI assistants and agents to read analytics and manage advertising across Google Ads, Google Analytics 4, and Meta Ads through natural language, with preview-confirm guardrails and configurable account and budget limits.21MIT