security(#300): trust-gate self-improve issues; sandbox build/test env#310
Draft
nolanmak wants to merge 2 commits into
Draft
security(#300): trust-gate self-improve issues; sandbox build/test env#310nolanmak wants to merge 2 commits into
nolanmak wants to merge 2 commits into
Conversation
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
|
Important Review skippedDraft detected. Please check the settings in the CodeRabbit UI or the ⚙️ Run configurationConfiguration used: defaults Review profile: CHILL Plan: Pro Run ID: You can disable this status message by setting the Use the checkbox below for a quick retry:
✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
…gress limitation Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
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 free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Closes #300.
Self-improve issue bodies are attacker-controllable (public repo) yet flowed raw into a Write/Edit/Bash reasoner and a host-side
cargo/npmgate that executesbuild.rs/proc-macros/npm postinstall. This adds two additive guards without removing the feature:authorAssociationOWNER/MEMBER/COLLABORATOR), or anAUGMENTAGENT_SELFIMPROVE_TRUSTED_AUTHORSlogin auto-select. Untrusted-authored issues are refused before any worktree, reasoner call, build, or branch push (owner must opt them in). Default-deny.env_clear()+ provider/API secrets stripped by name (OPENAI/ANTHROPIC/COMPOSIO/GH_TOKEN/…), so a hostile build script can't exfiltrate keys. Behavior-preserving — honest builds don't need those keys.is_blast_radiuskept as defense-in-depth. Owner-authored issues run end-to-end unchanged (minus secrets in the build child).Only
crates/augmentagent-cli/src/self_improve.rschanged (+337/−10).Verification:
cargo check -p augmentagent-clipasses.cargo test … self_improve= 7 pass / 1 fail, where the failure (workspace_overlapping_deploy_checkout_is_refused) also fails on unmodified HEAD~1 — a pre-existing/tmp-symlink test-isolation flake, not introduced here. 3 new tests pass.Honest limitation: network egress is not blocked in-process (no host mechanism); defense is layered (untrusted issues never reach the single-repo gate; multi-repo runs in an ephemeral clone behind owner approval). A container/network-namespace sandbox would close this fully — follow-up.
🤖 swarm-authored, human-review-required (draft).