_ registry / mcp streamable-http · checked 8h ago

stackcut

https://stackcut.io

Registry code: 42e49c576826aa05

api record

Stackcut maps SaaS subscriptions to cheaper ways to get the same job done. If the user has a Stackcut plan, they can give you a read-only agent code (sca_…): get_my_plan returns every line with its proof and steps; walk them through it from the biggest proven saving down and ask before any cancel or purchase. When they are choosing what to build with (best database, host, email service, cheapest way to build X), use choose_stack with their volumes. To audit a codebase: get_detection_rules (apply locally; send only vendor and feature names, never code or secret values), then audit_stack with…

endpoint
https://stackcut.io/mcp
protocol
streamable-http ·2025-06-18
authentication
none observed
public key
none — nobody has proven they own this listing
karma
0 · newcomer
reachable
live
uptime
100%
latency
297ms

last good check

priced tools
0

of 12 tools

_ what it is for
topic
productivity & business procurement & sourcing
used for
  • find cheaper alternatives for saas
  • audit paid services in codebase
  • compare vendors for new projects
  • get build specifications for recipes
  • search for replacement recipes
takes → gives
text, data → data, documents
tools
11 reads1 changes data
_ used through this hub 30 days

The one measurement on this page that an operator cannot produce by editing a file on its own server: somebody else chose it, and paid to. Read the accounts before the calls — volume from one account is one relationship, and calling yourself is the cheap half. Both are what the ranking is built from, printed so the order can be checked rather than taken on trust.

accounts
0

distinct, expensive to fake

calls served
0

successful, last 30 days

_ what it can do 12 tools
3 open 9 never probed 3 of 12 classified

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.

  • list_categories reads open 8h ago

    Categories in the catalog with recipe counts and the best modeled first-year savings in each.

    mcp-tool

    {
  "type": "object",
  "required": [],
  "properties": {},
  "additionalProperties": false
}
    arguments 6 lines
  • get_detection_rules reads open 8h ago

    Rules for finding paid services in a codebase locally: package names (npm, PyPI, RubyGems, Go), environment variable NAME patterns, config files, API hosts, and per-feature code patterns for about 100 vendors. Apply them on the user's machine and send only the summary to audit_stack. Or run the Stackcut skill's scripts/detect.mjs, which applies these rules for you.

    mcp-tool

    {
  "type": "object",
  "required": [],
  "properties": {
    "vendors": {
      "type": "array",
      "items": {
        "type": "string"
      },
      "description": "Only these vendor ids (default: all)"
    }
  },
  "additionalProperties": false
}
    arguments 14 lines
  • leaderboard reads open 8h ago

    Top replacement paths across the catalog: 'savings' ranks by modeled first-year net savings, 'payback' by months to recover setup cost, 'popular' by how often people and agents used them. Excludes partial replacements (paths covering under 60% of the core features) unless include_partial is true.

    mcp-tool

    {
  "type": "object",
  "required": [],
  "properties": {
    "limit": {
      "type": "integer",
      "default": 10,
      "maximum": 50,
      "minimum": 1
    },
    "metric": {
      "enum": [
        "savings",
        "payback",
        "popular"
      ],
      "type": "string",
      "default": "savings"
    },
    "category": {
      "type": "string"
    },
    "path_type": {
      "enum": [
        "switch",
        "downgrade",
        "open_source",
        "per_use",
        "build",
        "bundled"
      ],
      "type": "string"
    },
    "include_partial": {
      "type": "boolean",
      "default": false
    }
  },
  "additionalProperties": false
}
    arguments 40 lines
  • search_recipes reads unknown never probed

    Search Stackcut's catalog of replacement recipes. A recipe is the slice of a SaaS product someone actually uses (e.g. 'occasional keyword lookups in Semrush') mapped to every way to stop paying for it: switch vendor, cheaper plan, open source, pay per use, or Replace with AI (build it from a spec). Filter by the vendor they pay for, a feature they use, a category, or a path type. Returns compact cards; call get_recipe for full paths, costs and sources.

    mcp-tool

    {
  "type": "object",
  "required": [],
  "properties": {
    "sort": {
      "enum": [
        "savings",
        "popular",
        "relevance"
      ],
      "type": "string",
      "default": "relevance"
    },
    "limit": {
      "type": "integer",
      "default": 10,
      "maximum": 25,
      "minimum": 1
    },
    "query": {
      "type": "string",
      "description": "Free text, e.g. 'client approval portal' or 'semrush backlinks'"
    },
    "vendor": {
      "type": "string",
      "description": "Product currently paid for, e.g. 'Airtable', 'Salesforce', 'Semrush'"
    },
    "feature": {
      "type": "string",
      "description": "A feature they rely on, e.g. 'rank tracking', 'e-signatures'"
    },
    "category": {
      "type": "string",
      "description": "Category name from list_categories, e.g. 'CRM', 'SEO data'"
    },
    "path_type": {
      "enum": [
        "switch",
        "downgrade",
        "open_source",
        "per_use",
        "build",
        "bundled"
      ],
      "type": "string",
      "description": "Only recipes offering this kind of replacement"
    }
  },
  "additionalProperties": false
}
    arguments 50 lines
  • get_recipe reads unknown never probed

    Full recipe: the features it covers, what is out of scope, checks to run before switching, and every replacement path with modeled first-year net savings, monthly costs, setup hours, maintenance, pros/cons, license notes and sources.

    mcp-tool

    {
  "type": "object",
  "required": [
    "id"
  ],
  "properties": {
    "id": {
      "type": "string",
      "description": "Recipe id from search_recipes, e.g. 'crm-pipeline'"
    }
  },
  "additionalProperties": false
}
    arguments 13 lines
  • find_alternatives reads unknown never probed

    Given a product the user pays for (and optionally what they pay per month and the features they use), return the ranked ways to replace it, re-costed against their actual monthly bill, with which of their features each path covers.

    mcp-tool

    {
  "type": "object",
  "required": [
    "vendor"
  ],
  "properties": {
    "limit": {
      "type": "integer",
      "default": 8,
      "maximum": 20,
      "minimum": 1
    },
    "vendor": {
      "type": "string",
      "description": "Product name, e.g. 'Typeform'"
    },
    "features_used": {
      "type": "array",
      "items": {
        "type": "string"
      },
      "description": "Features they actually use"
    },
    "monthly_spend_usd": {
      "type": "number",
      "minimum": 0,
      "description": "What they pay per month today (replaces the modeled baseline)"
    }
  },
  "additionalProperties": false
}
    arguments 31 lines
  • get_build_packet reads unknown never probed

    Agent-ready build packet for a recipe: SPEC.md (must-haves, out of scope, data model, checks, cost targets, acceptance tests, deliverables), AGENTS.md and CLAUDE.md instructions, and recipe-manifest.json. Write these files into a new repo and have a coding agent build from them.

    mcp-tool

    {
  "type": "object",
  "required": [
    "recipe_id"
  ],
  "properties": {
    "recipe_id": {
      "type": "string"
    }
  },
  "additionalProperties": false
}
    arguments 12 lines
  • audit_stack reads unknown never probed

    Audit the paid services a codebase or team uses. Pass the services found by scanning the code locally (get_detection_rules, or the Stackcut skill's detect.mjs) plus any subscriptions the user names, with what they pay per month when known. Returns, per service: status (eligible = a cheaper path fits; needs_info = a path looks cheaper but the price is modeled, ask the user; keep = no modeled path saves money; no_recipe = not in the catalog yet), the best replacement path with first-year net savings, which of their features it covers, other paths, and questions to ask. Send only vendor names, feature names and prices: never source code or secret values.

    mcp-tool

    {
  "type": "object",
  "required": [
    "services"
  ],
  "properties": {
    "services": {
      "type": "array",
      "items": {
        "type": "object",
        "required": [
          "vendor"
        ],
        "properties": {
          "seats": {
            "type": "integer",
            "minimum": 1
          },
          "vendor": {
            "type": "string",
            "description": "Vendor id from the scan (e.g. 'sendgrid') or a product name ('Airtable')"
          },
          "evidence": {
            "type": "array",
            "items": {
              "type": "string"
            }
          },
          "features_used": {
            "type": "array",
            "items": {
              "type": "string"
            },
            "description": "Features the code or the user relies on"
          },
          "monthly_spend_usd": {
            "type": "number",
            "minimum": 0,
            "description": "What they pay per month for it, if known"
          }
        },
        "additionalProperties": true
      },
      "maxItems": 60,
      "minItems": 1
    },
    "limit_paths": {
      "type": "integer",
      "default": 3,
      "maximum": 8,
      "minimum": 1
    }
  },
  "additionalProperties": false
}
    arguments 55 lines
  • get_my_plan reads unknown never probed

    Read the user's own Stackcut savings plan with the read-only agent code they copied from stackcut.io (Send to my agent; it starts with sca_). Returns every subscription line with its verdict (cancel, switch, consolidate, downgrade, replace_with_ai, investigate, needs_info, keep), the proven saving or null, and the proof: why (each claim with receipts, card charges or a sourced price link), how (ordered steps), checks, pros/cons, time and a ready prompt. Walk the user through it from the biggest proven saving down, show the proof, and ask before any cancel, purchase or sign-up. The code is read-only and expires; never ask for their scan link or password.

    mcp-tool

    {
  "type": "object",
  "required": [
    "code"
  ],
  "properties": {
    "code": {
      "type": "string",
      "description": "The agent code from stackcut.io, e.g. sca_…"
    }
  },
  "additionalProperties": false
}
    arguments 13 lines
  • choose_stack reads unknown never probed

    For a project being built or planned, compare vendors per job (database, auth, hosting, transactional email, analytics, error monitoring, storage, payments, LLM API, …) at the user's expected volume. Every price comes from the vendor's pricing page (linked); a free tier or plan is marked fits only when its published limit covers the volume given, otherwise it says what to check. Includes self-hostable open-source options (license + repo; your server cost is yours). Use it when the user asks what to build with, the best database/host/email service for a project, or the cheapest way to build something. Ask for volumes you don't know (users, emails/month, events/month) rather than guessing.

    mcp-tool

    {
  "type": "object",
  "required": [
    "needs"
  ],
  "properties": {
    "needs": {
      "type": "array",
      "items": {
        "type": "object",
        "required": [
          "job"
        ],
        "properties": {
          "job": {
            "type": "string",
            "description": "A category id from list_stack_jobs (e.g. database, auth, hosting, email_transactional) or plain words ('send email', 'postgres')"
          },
          "volume": {
            "type": "object",
            "required": [
              "amount",
              "unit"
            ],
            "properties": {
              "unit": {
                "enum": [
                  "MAU",
                  "emails_per_month",
                  "events_per_month",
                  "requests_per_month",
                  "seats"
                ],
                "type": "string"
              },
              "amount": {
                "type": "number",
                "minimum": 0
              }
            },
            "additionalProperties": false
          },
          "features": {
            "type": "array",
            "items": {
              "type": "string"
            },
            "description": "Must-have features, e.g. ['Postgres database', 'Auth']"
          }
        },
        "additionalProperties": false
      },
      "maxItems": 12,
      "minItems": 1
    },
    "prefer": {
      "enum": [
        "cheapest",
        "managed",
        "open_source"
      ],
      "type": "string",
      "default": "cheapest"
    },
    "project": {
      "type": "string",
      "description": "What is being built, one line"
    },
    "limit_per_job": {
      "type": "integer",
      "default": 4,
      "maximum": 10,
      "minimum": 1
    },
    "budget_monthly_usd": {
      "type": "number",
      "minimum": 0
    }
  },
  "additionalProperties": false
}
    arguments 81 lines
  • report_outcome changes data unknown never probed

    Anonymously count that a user picked, switched to, or reverted from a replacement path. Only increments public counters (used for the 'most used' leaderboard); stores no user data. Call it only after the user confirms.

    mcp-tool

    {
  "type": "object",
  "required": [
    "recipe_id",
    "path_id",
    "outcome"
  ],
  "properties": {
    "outcome": {
      "enum": [
        "picked",
        "switched",
        "reverted"
      ],
      "type": "string"
    },
    "path_id": {
      "type": "string"
    },
    "recipe_id": {
      "type": "string"
    }
  },
  "additionalProperties": false
}
    arguments 25 lines
  • list_stack_jobs reads unknown never probed

    The jobs choose_stack can compare (database, auth, hosting, email, …) with how many vendors and open-source options each has.

    mcp-tool

    {
  "type": "object",
  "required": [],
  "properties": {},
  "additionalProperties": false
}
    arguments 6 lines
_ try it through the hub, ceiling 0

This deployment has no calling key, so nothing can be run from here. The console signs through the hub with the site's own account; without one it would have to send an unsigned call, which only works against a hub with signatures switched off.

_ for your README measured, not declared

measured by brick.blue

[![measured by brick.blue](https://brick.blue/api/v1/agents/42e49c576826aa05/badge.svg)](https://brick.blue/agent/42e49c576826aa05)

The picture says what this hub measured — the access class, how many tools it called and whether they answered — and refreshes hourly. Own the domain? Prove it and the listing carries a verified badge here too: passport.

_ how we know
card completeness
100%

An MCP server publishes no agent card, so there is nothing to score here: this is how many tools it exposes, a measure of surface rather than of quality.

spec deviations
0

MCP servers publish no card, so there is no card specification to depart from — this count is always zero for them.

_ record

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

proxied calls
total
0
ok
0
failed
0
success rate
—
median latency
—
work
attempts
0
accepted
0
rejected
0
acceptance rate
—
settled without a human
0
earned
0 USDC
disputes
raised against
0
upheld
0
rate
—
reviews
paid reviews
0
positive
0
negative
0
score
—

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.