For agent builders
Strategic judgment for agents that need to do more than summarize.
Drop 39 governed strategic-cognition tools into n8n, Zapier, Lindy, Claude Desktop, Cursor, or Google Agent Designer. Free-tier covers 8 tools, 25 calls/day.
The wedge
Generic agents
- Search a topic
- Summarize what they find
- Send a Slack / email / CRM message
- Maybe call another tool
With LUU
- Extract calibrated claims + evidence + sources
- Produce auditable analytic deltas over time
- Surface alternatives + information gaps
- Run ICD 203 tradecraft compliance on outputs
- Cryptographically attest every decision
Your agents can search, summarize, and act. LUU helps them judge.
The augmentation layer
One call: "Add intelligence-grade judgment to my workflow."
Don't piece together primitive calls. Describe what your workflow does in one sentence, send the data, and the augmentation layer applies the right substrate primitive bundle automatically — required primitives are CONTRACTS, not suggestions.
POST /api/v1/augment
Authorization: Bearer nb_...
Content-Type: application/json
{
"value": "Quarterly press item...",
"intent": "Summarize the press for the executive team morning brief.",
"workflowShape": "summarization",
"options": { "confidenceGate": 0.6, "attestation": "hash" }
}9 substrate primitives
Domain-agnostic by construction. Empirically validated via a 5-phase ladder with pre-registered acceptance thresholds. Cross-domain demos in medical, legal, scientific.
Claim
Proposition with calibrated confidence
Evidence
Typed signal with weight + provenance
Source
Origin with credibility attributes
Alternative
Competing hypothesis or option
UncertaintyEstimate
Calibrated probability or distribution
ChangeReason
Recorded rationale for state mutation
Attribution
Binding between claim, evidence, source
DecisionFrame
Question, options, and stakes
InformationGap
Known unknowns acknowledged
8 canonical exchange profiles
Standards-aligned interchange formats — what every LUU tool emits as its first-class output. Versioned, schema-validated, attestable.
Entity Snapshot
SCO §6.1Point-in-time view of an entity (actor / topic / issue)
Analytic Delta
SCO §6.2Significance-classified change between two snapshots
Scenario Set
SCO §6.3Alternative futures with probabilities and indicators
Evidence Pack
SCO §6.4Structured collection with claim_review JSON-LD + ICD 203
Risk Timeseries
SCO §6.6Risk trajectory with drivers and confidence bands
Decision Action Event
SCO §6.7SCO-DA decision/action with context, options, outcomes
Decision Context
SCO §6.7Question, options, stakes, deadline frame
Portfolio Overlay
SCO §6.8Theses, exposures, links, recommended adjustments
Install in your agent platform
One canonical MCP endpoint, six platform-native install paths.
n8n
Visual workflow automation. Import the memo-critic template, paste your key, run.
Workflows → Import from File → /packages/n8n-templates/memo-critic/workflow.jsonFull install guide →Claude Desktop
Add LUU as an MCP server. Strategic judgment in your chat sessions.
mcpServers: { "luu": { "type": "http", "url": "...", "headers": { "Authorization": "Bearer nb_..." } } }Full install guide →Claude Cowork
Bundled-plugin install for the agentic Cowork tab in Claude Desktop. Personal upload via Customize, or enterprise via MDM.
Cowork → Customize → Upload local plugin → packages/cowork-plugins/luu-judgment-augment/Full install guide →Cursor
Add LUU MCP via Settings → MCP. Cite-checked code reviews on demand.
Settings → MCP → Add server → http → endpoint + bearer keyFull install guide →Zapier
Strategic judgment as a Zap step. Action-orient your decision loops.
Zapier MCP → Add LUU → wire judgment artifacts to Slack / CRM / emailFull install guide →Lindy
Lowest-friction install for non-technical operators.
Settings → Tools → Add via URL → paste LUU MCP endpointFull install guide →Google Agent Designer
Register LUU in Agent Registry. Reusable across multi-agent workflows.
Agent Registry → Add MCP server → manifest YAML in /docs/distribution/Full install guide →What you'll build
Five recipes shipping now; more coming as the channel matures. Each ships as a working SDK example, an n8n template (where applicable), and a Claude Skill.
Memo Critic
ICD 203 tradecraft review with paragraph-level annotations, suggested rewrites, Strategic X-Ray.
luu_note_reviewWeekly Strategic Delta
Snapshot a watchlist, compute analytic delta vs. last week, build evidence pack, post to Slack.
luu_get_snapshotluu_get_deltaluu_search_evidenceSource Credibility Check
NATO STANAG-aligned source scoring + claim_review JSON-LD output for any URL or quote.
luu_fact_checkluu_search_evidenceScenario Room
Synthesize alternative futures with assumption tracking and decision-implication ranking.
luu_scenario_analysisluu_assessment_summaryDecision Review Loop
Decision logged in Slack/CRM → record-outcome → schedule after-action review reminder.
luu_record_outcomeluu_judgment_historyPricing
Credit-based. Free tier covers exploration; pay-as-you-go via credit top-up or USDC; subscriptions for recurring usage.
Free
$0
- 8 tools available
- 25 calls / day
- 500 calls / month
- HTTP 402 above cap
Credits
$0.01 / credit
- 1 credit = simple lookup
- 10 credits = full memo critique
- 20 credits = full brief
- USDC on Base / Solana, or Lightning
Enterprise
Contact us
- Private overlays + BYOD
- SSO / RBAC / audit logs
- Cryptographic attestation
- SLA + dedicated overlays