> For the complete documentation index, see [llms.txt](https://pingmexyz.gitbook.io/docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://pingmexyz.gitbook.io/docs/technical-resources/security-principles.md).

# Security Principles

* Secrets are never passed in raw form
* All inputs to public functions are hashes of secrets and salts
* On-chain inputs are public but uninformative without secrets
* Salts are created from email + password, and they evolve with each action to protect prior data
* Salts cannot be reused, and their seeds (such as email + password) are never disclosed
* To safeguard against front-runners that monitor mempool for proofs, we use the Commit-Reveal scheme for all proof-validations
  * During the Commit, send the commits contains mapping from the next commitment to hash(proof + salt + globalSalt)
  * Only after the commit is verified off-chain, then Reveal — the protected business function submits the proof, salt and commitment. After successful verification, the business function proceeds, and the commit is removed at the end
  * By the time the proof is revealed, the commit is already on-chain, front-runners cannot front-run both transactions at the same time
* 2FA security and non-repudiation
* N of M password recovery
* Non-reentrant where applicable, safeguarding against race conditions
* Proofs submitted during claims and withdrawals are not visible on the blockchain until the transaction is concluded. This precludes the possibility of front-running
* Furthermore, sanity checks are available, especially before a proof is submitted, to guarantee successful completion of business functions. This will further guarantee prevention of accidental leakage of proofs
* Administrative pause() and unpause() for emergency halting as a backstop. notPaused modifier for all state-changing functions