{
  "id": 20608183,
  "title": "Bitcoin Covenants Part 6: FE Covenants — Enforcing covenants through cryptographically restricted signing keys",
  "url": "https://urgent.news/2026/11/18/bitcoin-covenants-part-6-fe-covenants-enforcing-covenants-through",
  "topic": "finance",
  "section": "Finance & Markets",
  "published": "2026-11-18T14:04:50.000Z",
  "source": {
    "name": "Cointelegraph",
    "slug": "cointelegraph",
    "url": "https://cointelegraph.com/research/bitcoin-covenants-part-6-fe-covenants-enforcing-covenants-through-cryptographically-restricted-signing-keys"
  },
  "original_language": "en",
  "account": "FE covenants utilize signing keys that contain spending constraints, allowing only certain transactions to be signed. This is the sixth part of a technical article series on Bitcoin covenants by Cointelegraph Research. Achieving changes in Bitcoin's consensus rules is challenging, as it requires coordination among node operators, miners, developers, and economic actors. Researchers have been investigating ways to enforce covenant behavior without altering consensus rules. In May 2024, Jeremy Rubin proposed a solution in a paper titled \"FE'd Up Covenants.\" His design uses functional encryption (FE) to enforce covenant rules within the signing key, rather than modifying Bitcoin's Script. In traditional public-key cryptography, a private key can sign any message. With FE-based covenants, signing is restricted to predefined valid transactions. Bitcoin nodes would verify signatures using OP_CHECKSIG as they currently do, eliminating the need for a soft fork. Functional encryption generalizes public-key encryption, where a private key decrypts a ciphertext to reveal plaintext. In FE, a master secret generates subsidiary functional keys, each revealing a specific function of the plaintext. The holder of a functional key can compute f(x) for a transaction tx without learning the plaintext x. Functional encryption was first formalized in 2011 by Boneh, Sahai, and Waters, with constructions for arbitrary functions emerging in 2013. Practical implementations became possible in 2021 with the use of well-founded cryptographic assumptions. In Rubin's design, f represents the signing function, taking two inputs: a private key and the transaction being signed. This multi-input FE allows for the creation of covenants with varying spending policies. A setup authority generates a Bitcoin key pair (private key k and public key P) and an FE key pair (master secret msk and FE public key mpk). The setup authority also publishes P, mpk, k encrypted to mpk, and the functional key derived from msk. The private key is destroyed, ensuring it cannot be used to violate the covenant. New covenants can be created indefinitely by tweaking the initial key pair (k, P) with the spending policy. Covenant receiving addresses commit to a 2-of-2 script, requiring a signature from the user's key and the FE signature. This prevents unauthorized parties from triggering valid transactions. The general version of FE-based covenants with f(k, tx, program, proof) is more expressive than OP_CTV, allowing for complex conditions and data-dependent policies.",
  "summary": "FE covenants encode spending constraints into the signing key, so only transactions that satisfy a predefined condition can be signed.",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}