PortfolioFit

Measured conclusion: Deep-tested. L2 outcome: tools_ok.

Check normalized portfolios against mandates and calculate allocation constraints.

Add to your agent

Copy this into your MCP client config:

{
  "mcpServers": {
    "PortfolioFit": {
      "url": "https://portfolio.fit/mcp",
      "transport": "streamable-http"
    }
  }
}

Measured tools (10)

• list_analysis_portfolios read-only

  List owner-scoped portfolios, analytics datasets, and profile readiness.

  Hints: read-only yes · destructive no · idempotent yes · open-world no

• list_portfolio_series read-only

  Discover owner-scoped immutable series profiles and their exact dataset versions. This returns metadata and coverage only, never dataset rows.

  Parameters:

  • cursor
  • statuses
  • page_size — integer
  • portfolio_id

  Hints: read-only yes · destructive no · idempotent yes · open-world no

• list_analyses read-only

  Read private calculation groups and independent operations. Filter by kind, portfolio, dates, ready/running/attention status or legacy open/finished status.

  Parameters:

  • kind
  • page — integer
  • status
  • date_to
  • date_from
  • page_size — integer
  • portfolio_id

  Hints: read-only yes · destructive no · idempotent yes · open-world no

• get_analysis read-only

  Read retained normalized inputs, operations, results and owner-scoped dataset references. This is not an external agent transcript.

  Parameters:

  • analysis_id (required) — string

  Hints: read-only yes · destructive no · idempotent yes · open-world no

• evaluate_portfolio_alignment

  Compare a clean mandate with normalized holdings. Do not send PDF, CSV, spreadsheet, or factsheet files; extract them before calling.

  Parameters:

  • holdings (required) — array
  • idempotency_key (required) — string
  • rules
  • mandate — string
  • analysis_name — Optional private grouping name. The same case-sensitive name, after trimming outer whitespace, reuses one group for this user. Omit it for an independent operation. Keep the name on retries; each calculation still needs its own idempotency_key.
  • base_currency — string
  • portfolio_name — string
  • source_warnings
  • composition_scope — string

  Hints: read-only no · destructive no · idempotent yes · open-world no

• calculate_portfolio_constraints

  Calculate explicit portfolio constraints deterministically from normalized holdings. Use this for arithmetic checks, not for interpreting natural-language mandates.

  Parameters:

  • holdings (required) — array
  • constraints (required) — array
  • idempotency_key (required) — string
  • analysis_name — Optional private grouping name. The same case-sensitive name, after trimming outer whitespace, reuses one group for this user. Omit it for an independent operation. Keep the name on retries; each calculation still needs its own idempotency_key.
  • base_currency — string
  • source_warnings
  • composition_scope — string

  Hints: read-only no · destructive no · idempotent yes · open-world no

• upload_portfolio_series_rows

  Upload normalized series_key/date/value/distribution rows and commit a dataset version. For up to 5,000 rows, omit chunk_index and total_chunks to complete in one call. For larger inputs, declare total_chunks on every call, send chunks in order, and reuse upload_id and idempotency_key. The final chunk automatically verifies and commits all data and returns immutable series profiles. Parse files locally before calling.

  Parameters:

  • idempotency_key (required) — string
  • rows (required) — array
  • source
  • upload_id
  • dataset_id
  • chunk_index — integer — Zero-based chunk position; upload chunks in order.
  • content_hash — string
  • dataset_name — string
  • portfolio_id
  • total_chunks — integer — Total number of chunks in this upload. The last accepted chunk commits the dataset automatically.
  • analysis_name — Optional private grouping name. The same case-sensitive name, after trimming outer whitespace, reuses one group for this user. Omit it for an independent operation. Keep the name on retries; each calculation still needs its own idempotency_key.
  • series_profiles — Source-backed semantics for each series_key. Supply on the first or final chunk; the final chunk can complete the proposals before commit. Ambiguous or incomplete profiles require review before calculation.
  • new_portfolio_name — string

  Hints: read-only no · destructive no · idempotent yes · open-world no

• cancel_portfolio_series_upload destructive

  Cancel a pending upload and permanently remove its staging chunks.

  Parameters:

  • upload_id (required) — string
  • analysis_name — Optional private grouping name. The same case-sensitive name, after trimming outer whitespace, reuses one group for this user. Omit it for an independent operation. Keep the name on retries; each calculation still needs its own idempotency_key.

  Hints: read-only no · destructive yes · idempotent yes · open-world no

• review_portfolio_series_profile

  Create an immutable profile revision using revise, or explicitly confirm a warning-free proposal with confirmation=true.

  Parameters:

  • profile_id (required) — string
  • action (required) — string
  • revisions
  • confirmation — boolean
  • analysis_name — Optional private grouping name. The same case-sensitive name, after trimming outer whitespace, reuses one group for this user. Omit it for an independent operation. Keep the name on retries; each calculation still needs its own idempotency_key.

  Hints: read-only no · destructive no · idempotent no · open-world no

• calculate_portfolio_statistics

  Calculate historical return, volatility, drawdown, risk-adjusted and benchmark-relative statistics from a confirmed portfolio series profile. The metrics parameter describes all available calculations and their required inputs. Use daily, weekly or monthly complete periods; results include coverage, warnings and exact dataset/profile pins. Reuse an idempotency_key only for an identical calculation.

  Parameters:

  • series_profile_id (required) — string — Confirmed portfolio series profile returned by upload or list_portfolio_series; pins one immutable dataset version.
  • metrics (required) — array — Select one or more unique metric names. periodic_returns: Return for each complete period at target_frequency. cumulative_return: Compounded return across the selected complete periods. observation_count: Number of source observations in the selected date range. period_count: Number of complete return periods available for calculation. min_return: Lowest periodic return and its period end date. max_return: Highest periodic return and its period end date. volatility: Sample standard deviation of periodic returns, with annualized volatility. drawdown: Decline from the running peak of compounded wealth for each period. max_drawdown: Largest peak-to-trough decline, with peak, trough and recovery dates when available. downside_deviation: Root mean square shortfall below minimum_acceptable_return, with annualized deviation. Requires minimum_acceptable_return. sharpe: Mean return above risk_free_rate divided by the sample standard deviation of excess returns, with annualized ratio. Requires risk_free_rate. sortino: Mean return above minimum_acceptable_return divided by downside deviation, with annualized ratio. Requires minimum_acceptable_return. beta: Sensitivity of portfolio returns to benchmark returns. Requires benchmark_profile_id. alpha: Return above the benchmark return implied by beta and risk_free_rate, with annualized alpha. Requires benchmark_profile_id and risk_free_rate. correlation: Pearson correlation between portfolio and benchmark returns. Requires benchmark_profile_id. tracking_error: Sample standard deviation of portfolio-minus-benchmark returns, with annualized tracking error. Requires benchmark_profile_id. information_ratio: Mean portfolio-minus-benchmark return divided by tracking error, with annualized ratio. Requires benchmark_profile_id.
  • target_frequency (required) — string — Calculate complete daily, weekly or monthly return periods. Annualization uses 252, 52 or 12 periods per year respectively; finer-frequency data must be available.
  • idempotency_key (required) — string
  • end_date
  • start_date
  • analysis_name — Optional private grouping name. The same case-sensitive name, after trimming outer whitespace, reuses one group for this user. Omit it for an independent operation. Keep the name on retries; each calculation still needs its own idempotency_key.
  • risk_free_rate — Required for sharpe and alpha. Use a confirmed risk-free rate profile or a source-backed dated annualized scalar with currency, tenor, compounding and provenance. Do not substitute an assumed rate.
  • benchmark_profile_id — Confirmed benchmark series profile in the same currency and return basis as the portfolio. Required for beta, alpha, correlation, tracking_error and information_ratio. Returns are matched by exact period dates.
  • minimum_acceptable_return — Required for downside_deviation and sortino. Use a confirmed minimum acceptable return profile or a source-backed dated annualized scalar with currency, tenor, compounding and provenance.

  Hints: read-only no · destructive no · idempotent yes · open-world no

Publisher website

Measured by MCPYP probing. Untested fields are omitted, never estimated. Listing is always free; MCPYP never proxies calls or stores user tokens.