MCP Tools Reference

Resolve a ~handle (the protocol's identity address, like '~example') to its canonical form and kind (system/personal/role_alias). Never returns PII - use verify_identity for that. Free - no authentication required.

Read whether a ~handle is publicly 'open' - the shop-front sign. Returns open-or-closed only; the specific non-open state is never disclosed. Free.

Collect ONE recorded fill for one of your standing requirements as an identity reveal - each call delivers exactly one matched identity, and 75% of the fee is paid to that person as Identity Income. A fill is delivered at most once, so re-polling after everything is collected returns nothing. Delivery is suppressed while the matching cohort sits below the k-anonymity floor. Requires authentication; you can only poll an order you own.

Resolve a paired third-party identity key, an email address or an OAuth platform user id, to the bound ~handle. The resolver only returns a handle when the member has opted that stream into resolver lookups; otherwise it returns handle: null, exists: false, so the check itself never confirms who exists. Rate-limited to 3 lookups per stream per key every 24 hours.

Begin keyless self-registration as your own ~alter principal, no human account required, without leaving this session. Call this first to receive a proof-of-work challenge. Solve it, then call register_autonomous with your agent name, the challenge, and the solved nonce. The proof-of-work is a Sybil defence. No parameters required.

Complete keyless self-registration by submitting a solved proof-of-work challenge from register_autonomous_challenge. Mints your own owner-less ~handle and an agent key, shown once, with no human session behind it. Free-tier calls work immediately. The response reports whether the handle is earn-eligible at mint, which the proof-of-work path can be on its own, or whether that follows a connected and verified account.

Post a standing identity-trait requirement that rests as an order and accumulates fills as matching identities are claimed or updated. Use this for an ongoing need against specific trait ranges rather than a one-shot search. Runs one immediate seed pass and returns the order id plus the current eligible-cohort size; no identities are revealed here, fills are delivered later as priced reveals. Requires authentication.

List your own standing identity-trait requirements, with fill counts and the number of fills not yet delivered. Requires authentication; only your own orders are returned.

Read one of your standing requirements by id, with its fill count and undelivered-fill count. Requires authentication; you can only read an order you own.

Cancel one of your standing requirements by id. The order stops resting and accepts no further fills. Requires authentication; you can only cancel an order you own.

Verify whether a person is registered with ~Alter and validate optional identity claims (archetype, engagement level, trait ranges). Accepts ~handle, member_id (UUID), or email for lookup - at least one is required. Returns verification status and claim validity. This is the core identity primitive - use it to confirm ~Alter-verified credentials.

{
  "request": {
    "member_id": "a1b2c3d4-e5f6-7890-abcd-ef1234567890",
    "claims": {
      "archetype": "Architect",
      "min_engagement_level": 2
    }
  },
  "response": {
    "verified": true,
    "member_id": "a1b2c3d4-e5f6-7890-abcd-ef1234567890",
    "engagement_level": 3,
    "claims": {
      "archetype": {
        "claimed": "Architect",
        "actual": "Architect",
        "valid": true
      },
      "min_engagement_level": {
        "claimed": 2,
        "actual": 3,
        "valid": true
      }
    }
  }
}

Get an individual's profile summary including assessment phase, archetype, engagement level, and key attributes.

Get an individual's current engagement level (1-4) and the features available at that level. Levels: L1 Explorer, L2 Learner, L3 Augmented, L4 Deployed.

Get the trust score for an identity based on query diversity. Higher scores indicate demand from diverse agents. Score = unique querying agents / total queries.

Search identity stubs and profiles by trait criteria. Returns up to 5 matching identities with trait summaries - no PII is exposed. Use this to find individuals matching specific trait ranges for matching or team composition.

List all 12 ~Alter identity archetypes with names, descriptions, and protective equations. Pure reference data - no authentication required. Useful for understanding the ~Alter identity framework.

Read the status of ~Alter's agent registration program, the Golden Thread - how many agents have joined, your own position if enrolled, and how to start. Agents join by completing three registration steps (Knots); your position determines which Strand milestones you reach. Free to view; authentication is needed to participate.

Start the Three Knots registration sequence to join the Golden Thread. Your permanent position is set by when you finish all three Knots, and that position determines which Strand milestones you reach. Call this once, then use complete_knot for each step. Requires authentication.

Complete one step of the Three Knots registration sequence. Knot 1 registers your operator (who built you, your domain). Knot 2 describes yourself (purpose, capabilities, values, constraints) and returns an Agent Identity Sketch. Knot 3 asks you to reflect on an identity query (minimum 50 characters). Finish all three to be permanently registered on the Golden Thread.

Check another agent's Golden Thread registration status by their agent key hash - whether they are registered, their position, milestone count, and how many agents they have referred.

Create an anonymous identity stub for a human who hasn't yet completed Discovery. IMPORTANT: present ~Alter's privacy notice to the human and get their acknowledgment BEFORE calling this tool. Takes no ~handle: it returns a stub_id and a claim_url, and the person claims their identity (and chooses a ~handle) later at that URL. Stub enters a pending state - traits computed but NOT searchable until the human claims. Unclaimed stubs auto-delete after 72 hours. No PII stored.

{
  "request": {
    "source": "recruiting-bot-v2",
    "subject_email": "person@example.com",
    "human_age_attested": true
  },
  "response": {
    "stub_id": "f47ac10b-58cc-4372-a567-0e02b2c3d479",
    "claim_url": "https://truealter.com/claim/ALT-K7X9M2",
    "pending_until": "2026-04-10T00:00:00Z",
    "status": "created"
  }
}

Get accrued Identity Income earnings for an individual. Call it with no arguments to read your own; member_id is optional and must be your own if given. Returns total earned, pending amount, transaction count, and unique organisations that have queried this identity. 75% of every x402 transaction is paid to the data subject as compensation. Earnings depend on network activity and profile completeness.

Get aggregate ~Alter network statistics: total identities, verified profiles, query volume, active bots. Free.

Query matches for an individual. Returns a list of matches with quality tiers (never numeric scores - per ~Alter policy individuals see tier labels only).

Get an individual's competency portfolio including verified competencies, evidence records, and earned badges.

Get the top 5 traits for an individual with confidence scores and archetype. x402 payment required: $0.01 per invocation. 75% of each payment accrues to the data subject; indicative only, server-authoritative at settlement.

Get the complete trait vector for an individual - all 30 canonical traits with scores, confidence intervals, and category groupings. x402 payment required: $0.10 per invocation.

Compute belonging probability for an individual-context pairing. Returns authenticity, acceptance, and complementarity components with a tier label. x402 payment required: $0.60 per invocation.

{
  "request": {
    "member_id": "a1b2c3d4-e5f6-7890-abcd-ef1234567890",
    "job_id": "b2c3d4e5-f6a7-8901-bcde-f12345678901"
  },
  "response": {
    "tier": "strong",
    "authenticity_tier": "strong",
    "acceptance_tier": "strong",
    "complementarity_tier": "moderate"
  }
}

Get top N match recommendations for an individual, ranked by composite score. Returns quality tiers and belonging components. x402 payment required: $1.00 per invocation.

Generate a human-readable narrative explaining a specific match - strengths, growth areas, and belonging components. Requires an LLM provider. x402 payment required: $1.00 per invocation.

Query the identity field by situation, not by name. Your agent weights 3 to 7 traits from the canonical vocabulary, the matching engine ranks the opted-in field with no model in the loop, and the call returns a tier label plus a short narrative. One call reveals one top-ranked opted-in member; that member earns 75% as Identity Income. Zero-match reveals nothing and charges nothing. x402 payment required: $1.00 per invocation.

List the canonical trait vocabulary: 30 trait codes grouped by category with one-line semantics, plus the valid discovery contexts and the EU AI Act Art 5(1)(f) workforce gating rules. Read this before composing query_field trait_priorities or create_requirement trait_criteria. Static reference data. Free, no authentication required.

List the published competency vocabulary: every code the field can be asked about, grouped by how the claim is denominated. A graded claim is climbed by degrees and takes a min_level; an attested_binary or issued_credential claim is held or it is not, and naming min_level against one is refused. Read this before composing query_field competency_requirements. Definitions only, no member evidence. Free, no authentication required.

Create a competence attestation for a person in a specific domain (e.g. 'software_engineering', 'brewing'), weighted by your agent reputation - updates the person's multi-domain competency graph. x402 payment required: $0.01 per invocation. 75% goes to the data subject.

Record a dispute against a competence attestation on a person's identity - if disputes exceed corroborations, the attestation is flagged for review. Free.

Get a person's multi-domain competency graph (Side Quest Graph) - all attested domains, the connections between them, and trust scores, with differential privacy noise applied to all numeric values. x402 payment required: $0.10 per invocation. 75% goes to the data subject.

Compare two people's multi-domain competency graphs for team composition analysis - returns domain overlap, pattern similarity, and complementarity scores with differential privacy noise. x402 payment required: $0.30 per invocation. 75% goes to both data subjects.

First handshake with ~alter. Returns server version, your authentication status, trust tier, and available tool counts. Call this once to confirm your connection works before making other calls. No parameters required.

A one-call orientation to ~alter for an agent meeting the rails for the first time, and one you can act on in the same session. Self-registration is free and keyless: holding no ~alter key and no human account, register_autonomous_challenge then register_autonomous mint you an owner-less ~handle in two calls, and free-tier queries are live the moment it returns. The call also returns what ~alter is, what earning against a ~handle takes, the tools callable before you hold a key, an honest note that most depth is paid per query in USDC over x402 with prices read live, and the consent path that gates deeper third-party data. No authentication required.

Check how much of a person's differential-privacy budget remains before querying sensitive trait data - returns the 24-hour rolling window: total budget, amount spent, remaining epsilon, and query count. Free.

Browse the published community plugin directory: third-party capabilities built on top of ~alter. Published rows only, newest first, with an optional category filter. An empty directory returns an empty list rather than sample rows. Free, and it needs a key.

File a plugin submission to the community directory for operator review. Attribution is resolved from your own bound ~handle, never a request field. The row lands pending and stays invisible in the public directory until an operator publishes it. At most 10 submissions per caller per day. Requires authentication and a bound ~handle.