fix(security): resolve CodeQL code-scanning alerts (XSS, biased RNG, workflow permissions) - #13
Merged
Conversation
Address high-confidence findings from CodeQL Advanced scanning: Reflected XSS (CWE-79) in OAuth callback error pages — HTML-escape the attacker-controllable `error`/`error_description` query params before interpolating into the response body: - packages/opencode/src/mcp/oauth-callback.ts - packages/opencode/src/plugin/openai/codex.ts - packages/opencode/src/plugin/snowflake-cortex.ts Biased random from a cryptographically secure source (CWE-330) — replace modulo reduction over a non-power-of-divisor alphabet with rejection sampling so each character is uniformly distributed: - packages/opencode/src/id/id.ts (base62) - packages/core/src/id/id.ts (base62) - packages/core/src/util/identifier.ts (base62) - packages/opencode/src/plugin/xai.ts (66-char PKCE alphabet) - packages/opencode/src/plugin/snowflake-cortex.ts (PKCE) - packages/opencode/src/plugin/openai/codex.ts (PKCE verifier) Workflow missing permissions — add least-privilege `contents: read`: - .github/workflows/notify-discord.yml - .github/workflows/storybook.yml - .github/workflows/typecheck.yml https://claude.ai/code/session_01EQ6cVmvSwQyDjsV8bkzwFP
This PR doesn't fully meet our contributing guidelines and PR template. What needs to be fixed:
Please edit this PR description to address the above within 2 hours, or it will be automatically closed. If you believe this was flagged incorrectly, please let a maintainer know. |
Thanks for your contribution! This PR doesn't have a linked issue. All PRs must reference an existing issue. Please:
See CONTRIBUTING.md for details. |
daemon-blockint-tech
marked this pull request as ready for review
June 14, 2026 07:48
Uh oh!
There was an error while loading. Please reload this page.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for freeto join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Issue for this PR
Closes #
Note: this repository has GitHub Issues disabled, so a tracking issue could not be opened. This PR resolves alerts surfaced directly by CodeQL Advanced code scanning on
dev.Type of change
What does this PR do?
Fixes high-confidence CodeQL Advanced code-scanning alerts on
dev. Each change is minimal and behavior-preserving.Reflected XSS — CWE-79 (alerts #14, #15, #16): the OAuth callback error pages interpolated the attacker-controllable
error/error_descriptionquery params straight into the HTML response. A victim driven to the local loopback callback with a craftederror_descriptionwould get arbitrary JS executed in their browser. Fixed by HTML-escaping the value before interpolation inoauth-callback.ts,plugin/openai/codex.ts, andplugin/snowflake-cortex.ts.Biased random from a CSPRNG — CWE-330 (alerts #33–#38): the ID and PKCE generators reduced random bytes with
% nover alphabets of length 62 and 66, neither of which divides 256, so the first few characters were over-represented. Replaced with rejection sampling (discard bytes in the biased tail) so every character is uniformly distributed — inid/id.ts(opencode + core),core/util/identifier.ts, and thexai/snowflake-cortex/openai/codexPKCE generators. I understand why this works: 256 = 4·62 + 8, so values ≥ 248 (resp. the tail for 66) map to a smaller residue range and skew the distribution; rejecting that tail removes the skew.Workflow missing permissions (alerts #1, #2, #3): added least-privilege top-level
permissions: contents: readtonotify-discord.yml,storybook.yml, andtypecheck.yml.Remaining CodeQL alerts (Critical SSRF / uncontrolled command line, dev-script sanitization, regex complexity, test/generated-file findings) are intentionally left for separate follow-up as they need deeper design review.
How did you verify your code works?
Screenshots / recordings
N/A — no UI changes.
Checklist
https://claude.ai/code/session_01EQ6cVmvSwQyDjsV8bkzwFP