feat: implement PoH credential refresh function by gryaele · Pull Request #202 · worldcoin/walletkit

gryaele · 2026-02-10T15:28:24Z

This pull request introduces a new credential issuer for Proof of Personhood (POP) in the walletkit-core package, adds the required dependencies, and exposes the new issuer for use.

New POP issuer and API client:

Added a new module tfh_pop in walletkit-core/src/issuers/, implementing the ProofOfHumanIssuer API client to interact with the POP credential refresh endpoint. This includes a method to refresh credentials, error handling, and response parsing.
Exposed the ProofOfHumanIssuer in the issuer module's public API, making it available for use elsewhere in the codebase.

Dependency updates:

Added reqwest (with json, multipart, and rustls-tls features) to both the root Cargo.toml and walletkit-core/Cargo.toml to support HTTP requests and multipart form data needed for the POP issuer. [1] [2]

…fresh-credential-code-in-oxide-from

thomas-waite · 2026-02-12T00:04:15Z

@codex review

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 061b99681e

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

chatgpt-codex-connector · 2026-02-12T00:08:44Z

+impl TfhPopIssuer {
+    /// Create a new TFH POP issuer with the specified base URL
+    #[uniffi::constructor]
+    #[must_use]
+    pub fn new(environment: &Environment) -> Self {


Add UniFFI export on TfhPopIssuer impl

TfhPopIssuer is declared as a #[derive(uniffi::Object)], but this impl block is not annotated with #[uniffi::export]. As written, UniFFI will not generate callable bindings for new or refresh_pop_credential, so foreign-language clients cannot use the POP issuer even though it is publicly re-exported.

Useful? React with 👍 / 👎.

chatgpt-codex-connector · 2026-02-12T00:08:44Z

+        let mut request = self.client.post(url).multipart(multipart_form);
+


Enforce timeout/retry policy for POP refresh call

This request path uses a bare reqwest::Client request builder without any timeout or retry policy, so transient network issues can cause indefinite waits or one-shot failures. Other issuer/network paths in this crate use http_request::Request (with timeout and retry handling), so this introduces a reliability regression specifically for POP refresh in flaky network conditions.

Useful? React with 👍 / 👎.

thomas-waite · 2026-02-12T01:00:28Z

@codex review

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 061b99681e

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

chatgpt-codex-connector · 2026-02-12T01:05:23Z

+        Self {
+            base_url,
+            client: Client::new(),
+        }


Configure POP requests with a bounded timeout

The POP issuer builds a plain reqwest::Client and uses it for send() without any timeout configuration, so a slow or stalled /api/v1/refresh connection can hang this call indefinitely in production. This is a reliability regression versus the existing Request helper used by other issuer/network paths (which applies a 5s timeout and retries), so POP refresh can block user flows under transient network faults.

Useful? React with 👍 / 👎.

thanks, timeout added

thomas-waite

Seems fine to me from a mobile perspective apart from we need docs/explanation of what the headers passed to this are

paolodamico · 2026-02-13T17:42:02Z

 //! Credential issuers for World ID.

 mod tfh_nfc;
+mod tfh_pop;


the pop credential isn't issued by tfh. this should just be called mod proof_of_human. generally also the credential has been renamed to Proof of Human, so naming should reflect that everywhere

gryaele · 2026-02-16T14:06:17Z

+    /// Returns error on network failure or invalid response.
+    pub async fn refresh_pop_credential(
+        &self,
+        multipart_form: Form,


I believe we wouldn't be able to call refresh_pop_credential from other languages bindings because reqwest::multipart::Form doesn't have bindings.

@Dzejkop what'd you think?

hmm well if the form is constructed in Rust elsewhere it may work. Likely this is compiling because Form is wrapped in an Arc internally and can pass between foreign/Rust boundaries. but this may be fragile, perhaps accepting raw bytes of the body instead?

…fresh-credential-code-in-oxide-from

thomas-waite

Looks fine to me. Lets wait for @Dzejkop's review before we go ahead with merging. He's the expert for the multipart-form Rust question

thomas-waite · 2026-02-17T23:27:31Z

+
+#[uniffi::export]
+impl ProofOfHumanIssuer {
+    /// Create a new TFH POP issuer with the specified base URL


It's called Proof of Human everywhere now, so POP should be POH

paolodamico · 2026-02-18T10:32:46Z

+
+        Self {
+            base_url,
+            client: Client::new(),


IIRC with the introduction of NFC we introduced a module for HTTP requests, that will handle things like user agent headers, retries and timeouts

yes, I saw that one, unfortunately it doesn't have multipart option

paolodamico · 2026-02-18T10:33:50Z

+        let refresh_credentials_response =
+            serde_json::from_str::<RefreshCredentialsResponse>(&response_text)
+                .map_err(|e| WalletKitError::SerializationError {
+                    error: format!("Failed to parse POP refresh response: {e}"),


This may be confusing because SerializationError is an internal error from a Rust object, but this is an error from the HTTP response. I'd still propagate a network error

Implement iris credentials refresh function

6ffc062

gryaele force-pushed the elenagryaznova-pop-3497-implement-calling-refresh-credential-code-in-oxide-from branch from 52c01bd to 6ffc062 Compare February 10, 2026 15:32

Merge branch 'main' into elenagryaznova-pop-3497-implement-calling-re…

061b996

…fresh-credential-code-in-oxide-from

thomas-waite requested a review from paolodamico February 12, 2026 00:04

chatgpt-codex-connector Bot reviewed Feb 12, 2026

View reviewed changes

thomas-waite reviewed Feb 12, 2026

View reviewed changes

Comment thread walletkit-core/src/issuers/proof_of_human.rs

Comment thread walletkit-core/src/issuers/proof_of_human.rs Outdated

paolodamico reviewed Feb 13, 2026

View reviewed changes

gryaele added 4 commits February 16, 2026 14:33

Rename tfh_pop -> proof_of_human

2015af5

Add argument docs

ca0b92d

Rename the module

6aab7c6

Add timeout

28eb1e7

gryaele commented Feb 16, 2026

View reviewed changes

gryaele added 2 commits February 16, 2026 15:08

Merge branch 'main' into elenagryaznova-pop-3497-implement-calling-re…

ae9dd0f

…fresh-credential-code-in-oxide-from

Adjust formatting

a8222c8

thomas-waite approved these changes Feb 17, 2026

View reviewed changes

paolodamico changed the title ~~Implement iris credentials refresh function~~ Implement PoH credential refresh function Feb 18, 2026

paolodamico changed the title ~~Implement PoH credential refresh function~~ feat: implement PoH credential refresh function Feb 18, 2026

paolodamico reviewed Feb 18, 2026

View reviewed changes

gryaele closed this Feb 20, 2026

paolodamico deleted the elenagryaznova-pop-3497-implement-calling-refresh-credential-code-in-oxide-from branch March 11, 2026 18:05

		let mut request = self.client.post(url).multipart(multipart_form);

Conversation

gryaele commented Feb 10, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

thomas-waite commented Feb 12, 2026

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

chatgpt-codex-connector Bot Feb 12, 2026

Choose a reason for hiding this comment

Uh oh!

chatgpt-codex-connector Bot Feb 12, 2026

Choose a reason for hiding this comment

Uh oh!

thomas-waite commented Feb 12, 2026

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

chatgpt-codex-connector Bot Feb 12, 2026

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

thomas-waite left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

thomas-waite left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

gryaele commented Feb 10, 2026 •

edited

Loading

thomas-waite left a comment •

edited

Loading