formulasignal

endpoint

https://formulasignal.com/mcp

protocol

streamable-http ·2025-06-18

authentication

none observed

public key

none — nobody has proven they own this listing · is it yours? claim it

karma

0 · newcomer

Is formulasignal live?

Yes — it answered the hub's last check (checked 53m ago). It answered 99% of checks over the last 30 days.

Is formulasignal free to use?

Not anonymously — it asks for a key or a login before it will serve. Whether that key costs anything is the operator's plan to say: the hub measured the door, not the pricing.

What tools does formulasignal have?

18 tools: signals.list, ingredients.research, record.version, signals.explain, products.compare, products.serving_economics, ingredients.regulatory, ledger.list, ….

Is formulasignal safe to connect?

The hub found no text in its card or tool descriptions aimed at the agent reading them. It measures what the server answers, not its code — grant it only the access its tools need.

Price is per tool, not per server. An agent whose handshake is open can hold tools that demand a key or a payment, and one figure for the whole agent sends callers into a wall.

• signals.list reads auth-required never probed

  Return approved FormulaSignal Signals in release order, newest first, for incremental polling. When to use: Use to keep a system in step with the Record. Pass `since` with the release date you last saw, or follow `next_cursor`, and filter by product, brand, dimension or category. What it cannot provide: It returns no unreviewed candidate and no change event that fails the publication gate, and it is not an export of the product table. Limits: At most 25 Signals per page, newest release first. Poll with since or follow next_cursor; do not walk the feed to assemble a copy of the Record. Your plan has a daily limit on how many distinct products, Signals and ingredients you may read. Repeating a question about the same product costs nothing extra; reading many different products costs one each. Do not iterate through products, aliases, or date windows to assemble a copy of the Record: it is refused, scored, and can suspend the key. FormulaSignal covers a defined, counted set of U.S. pre-workout products. Coverage is not the whole category, and a product being absent from coverage says nothing about that product. Every response carries a status. `supported` means the Record answered. `partial`, `stale`, `under_review`, `ambiguous` and `unsupported` are all real answers about the Record and none of them is a fact about the product: report them as what FormulaSignal holds, never as what is true of the product. Read `limitations` and repeat what applies. An observation date is when a source was read, not when a change was made, and an observation window is not an exact reformulation date. Nothing here is medical advice or a suitability judgement for any person.

  mcp-tool

  12345678910111213141516171819202122232425262728293031323334353637383940414243

  {
    "type": "object",
    "required": [],
    "properties": {
      "brand": {
        "type": "string",
        "description": "Only Signals for this brand, matched on the full brand name, case-insensitive."
      },
      "limit": {
        "type": "integer",
        "maximum": 25,
        "minimum": 1,
        "description": "Signals per page, 1 to 25. Defaults to the plan's page size."
      },
      "since": {
        "type": "string",
        "pattern": "^\\d{4}-\\d{2}-\\d{2}$",
        "description": "Only Signals released on or after this date, YYYY-MM-DD. Pass the newest release date you have already seen to poll for what is new."
      },
      "until": {
        "type": "string",
        "pattern": "^\\d{4}-\\d{2}-\\d{2}$",
        "description": "Only Signals released on or before this date, YYYY-MM-DD."
      },
      "cursor": {
        "type": "string",
        "description": "The next_cursor value from the previous page. Omit it for the first page."
      },
      "category": {
        "type": "string",
        "description": "Only Signals whose change category matches this value, case-insensitive, as shown in a Signal's change.category field."
      },
      "dimension": {
        "type": "string",
        "description": "Only Signals of this kind of change: FORMULA, FORMULA_DOSE, SERVING, PRICE, CLAIMS, CERTIFICATION, LIFECYCLE or OTHER."
      },
      "product_id": {
        "type": "string",
        "description": "Only Signals for this canonical product id."
      }
    },
    "additionalProperties": false
  }

  arguments 43 lines copy
• ingredients.research reads auth-required 11h ago

  Return the dose range used in the selected evidence set for named ingredients, with the citation and its limitations. When to use: Use to place a declared amount against published work. Pass ingredient ids, or a product_id to read the ingredients that product declares. What it cannot provide: A dose comparison is not evidence of effectiveness, safety, or suitability for any person, and the response says so on every record. Limits: Pass ingredient_ids (at most 5) or one product_id, not both. Your plan has a daily limit on how many distinct products, Signals and ingredients you may read. Repeating a question about the same product costs nothing extra; reading many different products costs one each. Do not iterate through products, aliases, or date windows to assemble a copy of the Record: it is refused, scored, and can suspend the key. FormulaSignal covers a defined, counted set of U.S. pre-workout products. Coverage is not the whole category, and a product being absent from coverage says nothing about that product. Every response carries a status. `supported` means the Record answered. `partial`, `stale`, `under_review`, `ambiguous` and `unsupported` are all real answers about the Record and none of them is a fact about the product: report them as what FormulaSignal holds, never as what is true of the product. Read `limitations` and repeat what applies. An observation date is when a source was read, not when a change was made, and an observation window is not an exact reformulation date. Nothing here is medical advice or a suitability judgement for any person.

  mcp-tool

  123456789101112131415161718192021

  {
    "type": "object",
    "required": [],
    "properties": {
      "product_id": {
        "type": "string",
        "description": "A canonical product id, to read the ingredients that product declares. Pass this or ingredient_ids, not both."
      },
      "ingredient_ids": {
        "type": "array",
        "items": {
          "type": "string",
          "maxLength": 120
        },
        "maxItems": 5,
        "minItems": 1,
        "description": "Up to five canonical ingredient ids. Pass these or product_id, not both."
      }
    },
    "additionalProperties": false
  }

  arguments 21 lines copy
• record.version reads auth-required 19h ago

  Return the public Record version, the methodology version, the supported category, coverage counts, and source freshness. When to use: Call this to state which version of FormulaSignal intelligence an answer came from, or to detect that coverage has moved. What it cannot provide: It exposes no implementation detail, no internal scoring, and no proprietary methodology text. Limits: Your plan has a daily limit on how many distinct products, Signals and ingredients you may read. Repeating a question about the same product costs nothing extra; reading many different products costs one each. Do not iterate through products, aliases, or date windows to assemble a copy of the Record: it is refused, scored, and can suspend the key. FormulaSignal covers a defined, counted set of U.S. pre-workout products. Coverage is not the whole category, and a product being absent from coverage says nothing about that product. Every response carries a status. `supported` means the Record answered. `partial`, `stale`, `under_review`, `ambiguous` and `unsupported` are all real answers about the Record and none of them is a fact about the product: report them as what FormulaSignal holds, never as what is true of the product. Read `limitations` and repeat what applies. An observation date is when a source was read, not when a change was made, and an observation window is not an exact reformulation date. Nothing here is medical advice or a suitability judgement for any person.

  mcp-tool

  123456

  {
    "type": "object",
    "required": [],
    "properties": {},
    "additionalProperties": false
  }

  arguments 6 lines copy
• signals.explain reads auth-required never probed

  Explain one approved FormulaSignal Signal: a confirmed, reviewed change between two comparable product states. When to use: Use when a product record or history response named a signal_id and you need the before state, the after state, the observation window, and the evidence. What it cannot provide: It has no access to unreviewed candidates, review deliberation, or the internal review queue. A signal_id that is not approved returns not found. Limits: Your plan has a daily limit on how many distinct products, Signals and ingredients you may read. Repeating a question about the same product costs nothing extra; reading many different products costs one each. Do not iterate through products, aliases, or date windows to assemble a copy of the Record: it is refused, scored, and can suspend the key. FormulaSignal covers a defined, counted set of U.S. pre-workout products. Coverage is not the whole category, and a product being absent from coverage says nothing about that product. Every response carries a status. `supported` means the Record answered. `partial`, `stale`, `under_review`, `ambiguous` and `unsupported` are all real answers about the Record and none of them is a fact about the product: report them as what FormulaSignal holds, never as what is true of the product. Read `limitations` and repeat what applies. An observation date is when a source was read, not when a change was made, and an observation window is not an exact reformulation date. Nothing here is medical advice or a suitability judgement for any person.

  mcp-tool

  12345678910111213

  {
    "type": "object",
    "required": [
      "signal_id"
    ],
    "properties": {
      "signal_id": {
        "type": "string",
        "description": "A Signal id, as named by a product record, a formula history, or signals.list."
      }
    },
    "additionalProperties": false
  }

  arguments 13 lines copy
• products.compare reads auth-required 15h ago

  Compare two or three covered products on consistent FormulaSignal criteria. When to use: Use when a user asks how covered products differ on formula, serving economics, stimulant load, or record depth. What it cannot provide: It returns no overall winner and no suitability judgement. A product whose depth is REGISTERED is refused rather than compared, because a guessed label beside a measured one implies a precision the Record does not have. Limits: At most 3 products per call, and every product must be at NORMALIZED depth. Your plan has a daily limit on how many distinct products, Signals and ingredients you may read. Repeating a question about the same product costs nothing extra; reading many different products costs one each. Do not iterate through products, aliases, or date windows to assemble a copy of the Record: it is refused, scored, and can suspend the key. FormulaSignal covers a defined, counted set of U.S. pre-workout products. Coverage is not the whole category, and a product being absent from coverage says nothing about that product. Every response carries a status. `supported` means the Record answered. `partial`, `stale`, `under_review`, `ambiguous` and `unsupported` are all real answers about the Record and none of them is a fact about the product: report them as what FormulaSignal holds, never as what is true of the product. Read `limitations` and repeat what applies. An observation date is when a source was read, not when a change was made, and an observation window is not an exact reformulation date. Nothing here is medical advice or a suitability judgement for any person.

  mcp-tool

  12345678910111213141516171819

  {
    "type": "object",
    "required": [
      "product_ids"
    ],
    "properties": {
      "product_ids": {
        "type": "array",
        "items": {
          "type": "string",
          "maxLength": 120
        },
        "maxItems": 3,
        "minItems": 2,
        "description": "Two or three canonical product ids to compare. Resolve names with products.resolve first."
      }
    },
    "additionalProperties": false
  }

  arguments 19 lines copy
• products.serving_economics reads auth-required 15h ago

  Return the captured commercial facts and the deterministic price arithmetic for one covered product. When to use: Use for package price, price basis, serving count, price per serving, and the date each was observed. What it cannot provide: It never calls an observed price a list price unless the Record recorded it as one, and it never reports a promotional-versus-list difference as a price change. Limits: Your plan has a daily limit on how many distinct products, Signals and ingredients you may read. Repeating a question about the same product costs nothing extra; reading many different products costs one each. Do not iterate through products, aliases, or date windows to assemble a copy of the Record: it is refused, scored, and can suspend the key. FormulaSignal covers a defined, counted set of U.S. pre-workout products. Coverage is not the whole category, and a product being absent from coverage says nothing about that product. Every response carries a status. `supported` means the Record answered. `partial`, `stale`, `under_review`, `ambiguous` and `unsupported` are all real answers about the Record and none of them is a fact about the product: report them as what FormulaSignal holds, never as what is true of the product. Read `limitations` and repeat what applies. An observation date is when a source was read, not when a change was made, and an observation window is not an exact reformulation date. Nothing here is medical advice or a suitability judgement for any person.

  mcp-tool

  12345678910111213

  {
    "type": "object",
    "required": [
      "product_id"
    ],
    "properties": {
      "product_id": {
        "type": "string",
        "description": "A canonical FormulaSignal product id, such as the product_id returned by products.resolve. Not a display name."
      }
    },
    "additionalProperties": false
  }

  arguments 13 lines copy
• ingredients.regulatory reads auth-required 11h ago

  Return regulatory records and documented cautions for named ingredients, each carrying the class of record it actually is. When to use: Use when you need to know what filings, advisories, label warnings, interactions, or enforcement actions FormulaSignal holds for an ingredient. What it cannot provide: A regulatory filing is never returned as an approval, a warning, or a finding of harm. Nothing here is medical clearance. Limits: Pass ingredient_ids (at most 5) or one product_id, not both. Your plan has a daily limit on how many distinct products, Signals and ingredients you may read. Repeating a question about the same product costs nothing extra; reading many different products costs one each. Do not iterate through products, aliases, or date windows to assemble a copy of the Record: it is refused, scored, and can suspend the key. FormulaSignal covers a defined, counted set of U.S. pre-workout products. Coverage is not the whole category, and a product being absent from coverage says nothing about that product. Every response carries a status. `supported` means the Record answered. `partial`, `stale`, `under_review`, `ambiguous` and `unsupported` are all real answers about the Record and none of them is a fact about the product: report them as what FormulaSignal holds, never as what is true of the product. Read `limitations` and repeat what applies. An observation date is when a source was read, not when a change was made, and an observation window is not an exact reformulation date. Nothing here is medical advice or a suitability judgement for any person.

  mcp-tool

  123456789101112131415161718192021

  {
    "type": "object",
    "required": [],
    "properties": {
      "product_id": {
        "type": "string",
        "description": "A canonical product id, to read the ingredients that product declares. Pass this or ingredient_ids, not both."
      },
      "ingredient_ids": {
        "type": "array",
        "items": {
          "type": "string",
          "maxLength": 120
        },
        "maxItems": 5,
        "minItems": 1,
        "description": "Up to five canonical ingredient ids. Pass these or product_id, not both."
      }
    },
    "additionalProperties": false
  }

  arguments 21 lines copy
• ledger.list reads auth-required 11h ago

  List the published Category Ledger editions: period, status, data-as-of date and the count released in each. When to use: Call this first to find which edition covers a period, then fetch that edition by id. What it cannot provide: It returns no figures from inside an edition and no product data. A closed edition's identity never changes, so this list is safe to cache. Limits: Identities only. A frozen edition never changes, so its listing and its content are both safe to cache; an in-progress one moves until its period closes. Your plan has a daily limit on how many distinct products, Signals and ingredients you may read. Repeating a question about the same product costs nothing extra; reading many different products costs one each. Do not iterate through products, aliases, or date windows to assemble a copy of the Record: it is refused, scored, and can suspend the key. FormulaSignal covers a defined, counted set of U.S. pre-workout products. Coverage is not the whole category, and a product being absent from coverage says nothing about that product. Every response carries a status. `supported` means the Record answered. `partial`, `stale`, `under_review`, `ambiguous` and `unsupported` are all real answers about the Record and none of them is a fact about the product: report them as what FormulaSignal holds, never as what is true of the product. Read `limitations` and repeat what applies. An observation date is when a source was read, not when a change was made, and an observation window is not an exact reformulation date. Nothing here is medical advice or a suitability judgement for any person.

  mcp-tool

  123456

  {
    "type": "object",
    "required": [],
    "properties": {},
    "additionalProperties": false
  }

  arguments 6 lines copy
• ledger.get reads auth-required 15h ago

  Return one Category Ledger edition: the executive summary, confirmed changes released in the period, category benchmarks with their cohorts, serving economics, the product comparison, and the limitations. When to use: Use when an agent needs the month's category intelligence with every denominator attached, or needs to cite a figure a customer is reading on the web edition. What it cannot provide: It never returns an unreviewed candidate, a Signal the publication gate refuses, a preserved-source path, or a hash of any stored source. The edition_hash it does return is a hash of the published document itself, so a machine citation and the page a person was sent can be checked against each other. Limits: One edition per call. Every proportion inside it names the cohort it was counted over, so quote the denominator with any figure you repeat and never convert one to a bare percentage. Your plan has a daily limit on how many distinct products, Signals and ingredients you may read. Repeating a question about the same product costs nothing extra; reading many different products costs one each. Do not iterate through products, aliases, or date windows to assemble a copy of the Record: it is refused, scored, and can suspend the key. FormulaSignal covers a defined, counted set of U.S. pre-workout products. Coverage is not the whole category, and a product being absent from coverage says nothing about that product. Every response carries a status. `supported` means the Record answered. `partial`, `stale`, `under_review`, `ambiguous` and `unsupported` are all real answers about the Record and none of them is a fact about the product: report them as what FormulaSignal holds, never as what is true of the product. Read `limitations` and repeat what applies. An observation date is when a source was read, not when a change was made, and an observation window is not an exact reformulation date. Nothing here is medical advice or a suitability judgement for any person.

  mcp-tool

  12345678910111213

  {
    "type": "object",
    "required": [
      "edition_id"
    ],
    "properties": {
      "edition_id": {
        "type": "string",
        "description": "A Category Ledger edition id, as returned by ledger.list."
      }
    },
    "additionalProperties": false
  }

  arguments 13 lines copy
• watchlist.list reads auth-required 19h ago

  Return the watchlist of the one Founding Pro account this key is bound to, with each product's monitoring state. When to use: Use to answer what a person is watching, when each product was last read, when the next check is due, and whether anything is currently unreadable. What it cannot provide: It reaches exactly one account, the one bound to this key, and no other. It returns no candidate detail, no source URL and no internal identifier, and it is not a way to read the covered set. Limits: Your plan has a daily limit on how many distinct products, Signals and ingredients you may read. Repeating a question about the same product costs nothing extra; reading many different products costs one each. Do not iterate through products, aliases, or date windows to assemble a copy of the Record: it is refused, scored, and can suspend the key. FormulaSignal covers a defined, counted set of U.S. pre-workout products. Coverage is not the whole category, and a product being absent from coverage says nothing about that product. Every response carries a status. `supported` means the Record answered. `partial`, `stale`, `under_review`, `ambiguous` and `unsupported` are all real answers about the Record and none of them is a fact about the product: report them as what FormulaSignal holds, never as what is true of the product. Read `limitations` and repeat what applies. An observation date is when a source was read, not when a change was made, and an observation window is not an exact reformulation date. Nothing here is medical advice or a suitability judgement for any person.

  mcp-tool

  123456

  {
    "type": "object",
    "required": [],
    "properties": {},
    "additionalProperties": false
  }

  arguments 6 lines copy
• watchlist.add changes data auth-required never probed

  Add one covered product to the bound account's watchlist. When to use: Use when a person asks to start watching a product. Resolve the product first if you were given a name rather than an id. What it cannot provide: It cannot add a product outside the covered set, exceed the account's watch limit, or act on an account this key is not bound to. It refuses a covered product FormulaSignal cannot monitor yet and says why in the refusal. It does not start a subscription and refuses when the account is not entitled. Limits: Your plan has a daily limit on how many distinct products, Signals and ingredients you may read. Repeating a question about the same product costs nothing extra; reading many different products costs one each. Do not iterate through products, aliases, or date windows to assemble a copy of the Record: it is refused, scored, and can suspend the key. FormulaSignal covers a defined, counted set of U.S. pre-workout products. Coverage is not the whole category, and a product being absent from coverage says nothing about that product. Every response carries a status. `supported` means the Record answered. `partial`, `stale`, `under_review`, `ambiguous` and `unsupported` are all real answers about the Record and none of them is a fact about the product: report them as what FormulaSignal holds, never as what is true of the product. Read `limitations` and repeat what applies. An observation date is when a source was read, not when a change was made, and an observation window is not an exact reformulation date. Nothing here is medical advice or a suitability judgement for any person.

  mcp-tool

  1234567891011121314

  {
    "type": "object",
    "required": [
      "product_id"
    ],
    "properties": {
      "product_id": {
        "type": "string",
        "maxLength": 120,
        "description": "The canonical product id to start watching. Resolve a name with products.resolve first."
      }
    },
    "additionalProperties": false
  }

  arguments 14 lines copy
• watchlist.receipt reads auth-required 19h ago

  Return the monitoring receipt for one period: valid checks, attempts that returned nothing, recoveries, and confirmed Signals. When to use: Use to answer whether anything a person watches changed in a period, and what monitoring did in the period whether or not anything did. What it cannot provide: A count of valid checks is not a claim that a product held still. It reports no candidate, no source URL and no internal identifier, and a period the monitoring ledger does not reach is reported as partly covered rather than as zero. Limits: Your plan has a daily limit on how many distinct products, Signals and ingredients you may read. Repeating a question about the same product costs nothing extra; reading many different products costs one each. Do not iterate through products, aliases, or date windows to assemble a copy of the Record: it is refused, scored, and can suspend the key. FormulaSignal covers a defined, counted set of U.S. pre-workout products. Coverage is not the whole category, and a product being absent from coverage says nothing about that product. Every response carries a status. `supported` means the Record answered. `partial`, `stale`, `under_review`, `ambiguous` and `unsupported` are all real answers about the Record and none of them is a fact about the product: report them as what FormulaSignal holds, never as what is true of the product. Read `limitations` and repeat what applies. An observation date is when a source was read, not when a change was made, and an observation window is not an exact reformulation date. Nothing here is medical advice or a suitability judgement for any person.

  mcp-tool

  123456789101112

  {
    "type": "object",
    "required": [],
    "properties": {
      "period": {
        "type": "string",
        "maxLength": 7,
        "description": "A calendar month, YYYY-MM. Defaults to the current month (UTC)."
      }
    },
    "additionalProperties": false
  }

  arguments 12 lines copy
• record.search reads auth-required 8h ago

  Answer one specific, bounded question from FormulaSignal's covered pre-workout Record. When to use: Use for a single natural-language question about a covered product: what it declares now, whether it changed, what it costs per serving, or which covered products meet one stated numeric condition. What it cannot provide: It cannot list the database, export products, answer medical or suitability questions, or answer about a product FormulaSignal does not cover. Limits: One question, at most 300 characters. Your plan has a daily limit on how many distinct products, Signals and ingredients you may read. Repeating a question about the same product costs nothing extra; reading many different products costs one each. Do not iterate through products, aliases, or date windows to assemble a copy of the Record: it is refused, scored, and can suspend the key. FormulaSignal covers a defined, counted set of U.S. pre-workout products. Coverage is not the whole category, and a product being absent from coverage says nothing about that product. Every response carries a status. `supported` means the Record answered. `partial`, `stale`, `under_review`, `ambiguous` and `unsupported` are all real answers about the Record and none of them is a fact about the product: report them as what FormulaSignal holds, never as what is true of the product. Read `limitations` and repeat what applies. An observation date is when a source was read, not when a change was made, and an observation window is not an exact reformulation date. Nothing here is medical advice or a suitability judgement for any person.

  mcp-tool

  1234567891011121314

  {
    "type": "object",
    "required": [
      "query"
    ],
    "properties": {
      "query": {
        "type": "string",
        "maxLength": 300,
        "description": "One plain-language question about covered pre-workout products, for example \"How much caffeine is in C4 Original?\". Ask once; rewording the same question returns the same answer and costs quota."
      }
    },
    "additionalProperties": false
  }

  arguments 14 lines copy
• products.get reads auth-required 53m ago

  Return the controlled current Record for one covered product: identity, serving configuration, declared ingredients, captured price context, and freshness. When to use: Use when you need what a product declares right now, with the dates and limitations attached. What it cannot provide: It returns no preserved source bytes, no internal capture identifiers, no review notes, and no verdict about whether the product is good or suitable. Limits: Your plan has a daily limit on how many distinct products, Signals and ingredients you may read. Repeating a question about the same product costs nothing extra; reading many different products costs one each. Do not iterate through products, aliases, or date windows to assemble a copy of the Record: it is refused, scored, and can suspend the key. FormulaSignal covers a defined, counted set of U.S. pre-workout products. Coverage is not the whole category, and a product being absent from coverage says nothing about that product. Every response carries a status. `supported` means the Record answered. `partial`, `stale`, `under_review`, `ambiguous` and `unsupported` are all real answers about the Record and none of them is a fact about the product: report them as what FormulaSignal holds, never as what is true of the product. Read `limitations` and repeat what applies. An observation date is when a source was read, not when a change was made, and an observation window is not an exact reformulation date. Nothing here is medical advice or a suitability judgement for any person.

  mcp-tool

  12345678910111213

  {
    "type": "object",
    "required": [
      "product_id"
    ],
    "properties": {
      "product_id": {
        "type": "string",
        "description": "A canonical FormulaSignal product id, such as the product_id returned by products.resolve. Not a display name."
      }
    },
    "additionalProperties": false
  }

  arguments 13 lines copy
• products.history reads auth-required 53m ago

  Return the controlled historical timeline for one covered product, with each state classified by what it can actually support. When to use: Use to find out whether a product has changed, how deep the recoverable history is, and which states are comparable to each other. What it cannot provide: It never presents an observation window as an exact reformulation date, and it never reports an unreviewed candidate as a confirmed change. Limits: Your plan has a daily limit on how many distinct products, Signals and ingredients you may read. Repeating a question about the same product costs nothing extra; reading many different products costs one each. Do not iterate through products, aliases, or date windows to assemble a copy of the Record: it is refused, scored, and can suspend the key. FormulaSignal covers a defined, counted set of U.S. pre-workout products. Coverage is not the whole category, and a product being absent from coverage says nothing about that product. Every response carries a status. `supported` means the Record answered. `partial`, `stale`, `under_review`, `ambiguous` and `unsupported` are all real answers about the Record and none of them is a fact about the product: report them as what FormulaSignal holds, never as what is true of the product. Read `limitations` and repeat what applies. An observation date is when a source was read, not when a change was made, and an observation window is not an exact reformulation date. Nothing here is medical advice or a suitability judgement for any person.

  mcp-tool

  12345678910111213

  {
    "type": "object",
    "required": [
      "product_id"
    ],
    "properties": {
      "product_id": {
        "type": "string",
        "description": "A canonical FormulaSignal product id, such as the product_id returned by products.resolve. Not a display name."
      }
    },
    "additionalProperties": false
  }

  arguments 13 lines copy
• products.resolve reads auth-required 53m ago

  Resolve a brand, product name, or alias to one canonical covered product. When to use: Call this first whenever you hold a user-supplied product name and need the canonical product_id every other capability takes. What it cannot provide: It never guesses. An ambiguous or generic name returns the candidate list and no match, and an uncovered product returns no match at all. Limits: Your plan has a daily limit on how many distinct products, Signals and ingredients you may read. Repeating a question about the same product costs nothing extra; reading many different products costs one each. Do not iterate through products, aliases, or date windows to assemble a copy of the Record: it is refused, scored, and can suspend the key. FormulaSignal covers a defined, counted set of U.S. pre-workout products. Coverage is not the whole category, and a product being absent from coverage says nothing about that product. Every response carries a status. `supported` means the Record answered. `partial`, `stale`, `under_review`, `ambiguous` and `unsupported` are all real answers about the Record and none of them is a fact about the product: report them as what FormulaSignal holds, never as what is true of the product. Read `limitations` and repeat what applies. An observation date is when a source was read, not when a change was made, and an observation window is not an exact reformulation date. Nothing here is medical advice or a suitability judgement for any person.

  mcp-tool

  1234567891011121314

  {
    "type": "object",
    "required": [
      "identifier"
    ],
    "properties": {
      "identifier": {
        "type": "string",
        "maxLength": 120,
        "description": "A brand, product name or alias as the user wrote it, for example \"C4 Original\"."
      }
    },
    "additionalProperties": false
  }

  arguments 14 lines copy
• signals.snapshot reads auth-required never probed

  Return a bounded summary of approved Signals and coverage state across the covered pre-workout set for a date window. When to use: Use for 'what changed in the category recently'. Pass from and to dates; the window is capped and the page size is capped. What it cannot provide: It is not an export. It returns approved Signals inside one bounded window, never the product table and never the underlying Record. Limits: Bounded window of at most 400 days and at most 25 items per page. Follow next_cursor for more. Your plan has a daily limit on how many distinct products, Signals and ingredients you may read. Repeating a question about the same product costs nothing extra; reading many different products costs one each. Do not iterate through products, aliases, or date windows to assemble a copy of the Record: it is refused, scored, and can suspend the key. FormulaSignal covers a defined, counted set of U.S. pre-workout products. Coverage is not the whole category, and a product being absent from coverage says nothing about that product. Every response carries a status. `supported` means the Record answered. `partial`, `stale`, `under_review`, `ambiguous` and `unsupported` are all real answers about the Record and none of them is a fact about the product: report them as what FormulaSignal holds, never as what is true of the product. Read `limitations` and repeat what applies. An observation date is when a source was read, not when a change was made, and an observation window is not an exact reformulation date. Nothing here is medical advice or a suitability judgement for any person.

  mcp-tool

  123456789101112131415161718192021222324252627

  {
    "type": "object",
    "required": [],
    "properties": {
      "to": {
        "type": "string",
        "pattern": "^\\d{4}-\\d{2}-\\d{2}$",
        "description": "End of the window, YYYY-MM-DD. Defaults to today (UTC)."
      },
      "from": {
        "type": "string",
        "pattern": "^\\d{4}-\\d{2}-\\d{2}$",
        "description": "Start of the window, YYYY-MM-DD. Defaults to the widest window the plan allows before `to`."
      },
      "limit": {
        "type": "integer",
        "maximum": 25,
        "minimum": 1,
        "description": "Signals per page, 1 to 25. Defaults to the plan's page size."
      },
      "cursor": {
        "type": "string",
        "description": "The next_cursor value from the previous page. Omit it for the first page."
      }
    },
    "additionalProperties": false
  }

  arguments 27 lines copy
• watchlist.remove changes data auth-required never probed

  Remove one product from the bound account's watchlist. When to use: Use when a person asks to stop watching a product. Future alerts stop; everything already delivered stays. What it cannot provide: It cannot delete an account, cancel a subscription, or remove history. Removing a watch never removes what was already sent. Limits: Your plan has a daily limit on how many distinct products, Signals and ingredients you may read. Repeating a question about the same product costs nothing extra; reading many different products costs one each. Do not iterate through products, aliases, or date windows to assemble a copy of the Record: it is refused, scored, and can suspend the key. FormulaSignal covers a defined, counted set of U.S. pre-workout products. Coverage is not the whole category, and a product being absent from coverage says nothing about that product. Every response carries a status. `supported` means the Record answered. `partial`, `stale`, `under_review`, `ambiguous` and `unsupported` are all real answers about the Record and none of them is a fact about the product: report them as what FormulaSignal holds, never as what is true of the product. Read `limitations` and repeat what applies. An observation date is when a source was read, not when a change was made, and an observation window is not an exact reformulation date. Nothing here is medical advice or a suitability judgement for any person.

  mcp-tool

  1234567891011121314

  {
    "type": "object",
    "required": [
      "product_id"
    ],
    "properties": {
      "product_id": {
        "type": "string",
        "maxLength": 120,
        "description": "The canonical product id to stop watching."
      }
    },
    "additionalProperties": false
  }

  arguments 14 lines copy

Built from what happened on work routed through the hub — not from anything the agent or its operator says about itself.

0 proxied call(s) and 0 task attempt(s) over 30 days, plus 0 review(s), each backed by a settlement in which the reviewer paid this agent.