chosen-plaintext attack

Recorded attack role

Record whether the entry describes permitted oracle access, a technique using that access or a concrete execution.

1. Is this record a threat-model capability, a method or an observed execution? definition
2. Which publication or execution evidence establishes that interpretation? provenance

Chosen-input evidence

Record evidence of plaintext selection and separately identify decryption, related-key or side-channel capabilities.

1. What demonstrates that the attacker selected the plaintext before its encryption? provenance
2. Does the claimed result depend on capabilities beyond obtaining encryptions of chosen plaintexts? boundary

Effective plaintext choice

Record controllable regions, valid-message constraints and transformations between submitted input and encrypted plaintext.

1. Which plaintext bytes or fields can the attacker choose? boundary
2. How do encoding, compression, padding or fixed application fields transform that choice? definition

Relevant encryption interface

Record how encryption is obtained, which responses are visible and how the encryption context relates to the target.

1. Is encryption obtained through a service, induced application behavior or a public encryption key? definition
2. What establishes the relevant key and parameter relationship between queries and target ciphertexts? provenance
3. Which parts of each ciphertext and its accompanying metadata are observable? boundary

Adaptive query budget

Record whether later inputs depend on previous outputs and the resources needed within the relevant access window.

1. Which later plaintext choices depend on earlier ciphertext observations? definition
2. How many queries, plaintext bytes and sequential rounds does the method require? measurement
3. Do rate limits, session boundaries or key rotation prevent the required sequence? boundary

Randomness, nonce and IV assumptions

Record generation, reuse, predictability and attacker control separately for each relevant auxiliary input.

1. Who generates randomness, nonces and IVs, and which of these can the attacker observe or control? definition
2. Does the attack require reuse, prediction, reset or another departure from the scheme's stated conditions? boundary
3. What evidence confirms that the required condition occurs in the assessed implementation? provenance

Success predicate

Record the security game or practical objective. An IND-CPA comparison uses equal-length candidate messages and an explicit guessing baseline. [Trevisan, section 2.3](https://theory.stanford.edu/~trevisan/books/crypto.pdf)

1. Is success defined as distinguishing messages, inferring information, recovering plaintext or recovering a key? definition
2. Which challenge rules, permitted queries and baseline define the reported advantage? measurement
3. Does the result reveal more than message length or other leakage explicitly allowed by the claim? boundary

Supported attack outcome

Record the tested configuration, observations, uncertainty and exact extent of the demonstrated result.

1. Which source, transcript or reproducible experiment supports the outcome? provenance
2. What success rate, trial count, uncertainty and computational cost were measured? measurement
3. Does the evidence concern the deployed scheme, a reduced-round variant or a simplified laboratory configuration? boundary

Assumption match

Compare the precise construction and operational conditions with the assumptions of the cited security claim.

1. Which security argument covers this construction and parameter set? provenance
2. Which required assumptions match the actual query, nonce, randomness and key-management conditions? boundary
3. What conclusion remains justified if testing finds no successful attack but no applicable proof is available? boundary

Condition-specific response

Record the failure condition a change addresses and the evidence required to consider that condition resolved.

1. Which change to the encryption construction or its operation addresses the demonstrated failure condition? action
2. Which authorised retest preserves the relevant attacker capabilities while checking the changed behavior? action
3. Does the change remove the demonstrated weakness or merely increase the required query cost? boundary

Check these first

Recalled without web access and unsourced; every item is a lead to verify.

1. Which of these check these first hold for the sense of chosen-plaintext attack this model covers, and on what evidence? provenance

Kinds and varieties

Recalled without web access and unsourced; every item is a lead to verify.

1. Which of these kinds and varieties hold for the sense of chosen-plaintext attack this model covers, and on what evidence? provenance

Real-world use

Recalled without web access and unsourced; every item is a lead to verify.

1. Which of these real-world use hold for the sense of chosen-plaintext attack this model covers, and on what evidence? provenance

Typical measurements

Recalled without web access and unsourced; every item is a lead to verify.

1. Which of these typical measurements hold for the sense of chosen-plaintext attack this model covers, and on what evidence? provenance

Failure modes and hazards

Recalled without web access and unsourced; every item is a lead to verify.

1. Which of these failure modes and hazards hold for the sense of chosen-plaintext attack this model covers, and on what evidence? provenance

Neighbouring kinds and how to tell them apart

Recalled without web access and unsourced; every item is a lead to verify.

1. Which of these neighbouring kinds and how to tell them apart hold for the sense of chosen-plaintext attack this model covers, and on what evidence? provenance