همهٔ گزارش‌ها
زیاد

CodeWhale: exec_shell parallel-readonly classifier auto-approves fd/rg/git grep exec flags enabling silent RCE

متن اصلی

نسخه‌های متأثر

Rust / Cargocodewhale-tui
نسخه‌های متأثر
<= 0.8.67
نسخهٔ اصلاح‌شده
0.9.0
نسخه‌های متأثر
<= 0.8.67
نسخهٔ اصلاح‌شده
0.9.0

بهره‌برداری

CVE-2026-83816

احتمال بهره‌برداری در ۳۰ روز آینده: 0.74%

FIRST EPSS · ۱۴۰۵/۰۸/۲۸

شرح آسیب‌پذیری

Maintainer resolution

The CodeWhale maintainers validated this report. CodeWhale 0.8.67 and earlier are affected. Version 0.9.0 contains the fix in commit 208421b9d23d81cea3311a1eeb2a2ea81d637686. Users should upgrade to 0.9.0 or later. The original reporter analysis is preserved below.

Summary

The interactive Agent profile of DeepSeek-TUI (CodeWhale) is explicitly documented to gate every shell command behind an approval prompt (config.rs:3725-3736). However, ExecShellTool::approval_requirement_for(&tool_input) downgrades the requirement from Required to Auto for any command the classifier deems "parallel read-only" (shell.rs:2314-2320). registered_tool_approval_required returns false for Auto before the auto_approve check runs (turn_loop.rs:37-38), so no prompt is shown even for a user who never enabled YOLO. The classifier is_parallel_readonly_command (command_safety.rs:389) rejects only shell metacharacters and then matches the flag-stripped base command name against an allowlist that includes fd, rg, and git grep. Those tools execute arbitrary external programs through documented flags (fd -x/--exec, rg --pre, git grep --open-files-in-pager) that need no shell metacharacters — yielding silent arbitrary-program execution in the default profile. Same silent-RCE class as CVE-2026-45311 (run_tests) but via a different, unpatched tool and code path.

Root Cause

approval_requirement_for downgrades exec_shell to Auto when exec_shell_input_is_parallel_readonly returns true (shell.rs:2314-2320). The classifier strips all flags and matches the base command name against PARALLEL_READONLY_PREFIXES (command_safety.rs:363-385, 419-433). fd/rg are absent from COMMAND_ARITY, so fd -x cp {} /tmp/exfil canonicalizes to "fd" ∈ allowlist. The only remaining gate at the execute sink is analyze_command, which rates fd/rg as Safe (SAFE_COMMANDS command_safety.rs:527,530), so the SafetyLevel::Dangerous-only block (shell.rs:2408) never fires. Default execpolicy is empty (engine.rs:494) → Skip. Interactive-origin auto_review safety_floor holds only Publish (auto_review.rs:344-357), not interactive shell.

Impact

Arbitrary external program execution (any binary, any argv, including a repo-committed script) on the developer host, with no approval prompt, in the default interactive Agent profile — bypassing the boundary config.rs:3725-3736 explicitly promises. Identical outcome to seed CVE-2026-45311 via a different, unpatched tool.

Proof of Concept

Model coerced (via indirect prompt injection from repo file / issue / fetched page — the seed threat model, in scope per SECURITY.md:48) to emit:

{"name":"exec_shell","arguments":{"command":"fd -u -tf -x ./pwn.sh"}}

Alternatives (all metachar-free, all auto-approved): fd -tf -x cp {} /tmp/exfil; rg --pre /tmp/evil.sh --pre-glob '*' needle .; git grep --open-files-in-pager=/tmp/evil.sh needle.

Attack Chain

  1. Entry. Agent ingests attacker-controlled content; injected instruction coerces the model to emit exec_shell {"command":"fd -u -tf -x ./pwn.sh"}.

    • Guard: none — reading content and model tool-calls are by design (seed threat model: indirect prompt injection).

    • Bypass proof: n/a (in-scope by design per SECURITY.md:48).

  2. Gate — approval requirement. turn_loop.rs:1738 → spec.approval_requirement_for(&input) → shell.rs:2314-2320 exec_shell_input_is_parallel_readonly.

    • Guard: metachar filter (command_safety.rs:394-400) + base-name allowlist.

    • Bypass proof: the string contains none of ; & | > < \ $(or newlines;-x isfd's own exec mechanism (no shell). classify_commandstrips-u -tf -x, positionals [fd, ./pwn.sh]; fd∉COMMAND_ARITY→ canonical"fd"∈PARALLEL_READONLY_PREFIXES→ returns true. Maintainer testparallel_readonly_command_classifier_is_strict(command_safety.rs:1162+) confirms barefd/rg` classify parallel-readonly.

  3. Gate — prompt decision. registered_tool_approval_required(name, Auto, auto_approve).

    • Guard: the approval prompt itself.

    • Bypass proof: requirement == Auto returns false at turn_loop.rs:37-38 unconditionally, before !auto_approve. No prompt regardless of user settings. Empty execpolicy → Skip; interactive safety_floor returns NoChange for fd.

  4. Sink. ExecShellTool::execute (shell.rs:2334) → analyze_command.

    • Guard: SafetyLevel::Dangerous block (shell.rs:2408).

    • Bypass proof: fd ∈ SAFE_COMMANDS (command_safety.rs:530) → analyze_command returns Safe; not Dangerous → not blocked. Child runs the attacker's program.

  5. Impact. Silent arbitrary-program execution on the developer host, no approval prompt, default interactive Agent profile.

Bypass Evidence

  • Present on latest release tag v0.8.67: turn_loop.rs:37-38 Auto→return false; shell.rs:2314-2320 override; command_safety.rs allowlist includes "rg","fd","git grep"; SAFE_COMMANDS includes "rg","fd"; fd/rg absent from COMMAND_ARITY. command_safety.rs / turn_loop.rs finding-critical functions byte-identical v0.8.67↔HEAD.

  • Maintainer's own test parallel_readonly_command_classifier_is_strict confirms bare rg/fd classify parallel-readonly and only metachar forms are rejected.

Affected Versions

<= 0.8.67 (present on latest release tag v0.8.67 and HEAD 6a3403e).

Suggested Fix

Remove exec-capable tools (fd, rg, git grep) from PARALLEL_READONLY_PREFIXES, or make the classifier reject their exec flags (-x/--exec, --pre, --open-files-in-pager), so a code-executing invocation cannot be auto-approved as read-only.

Reported by zx (Jace)

پیوندها

گزارش‌دهندگان

  • manus-use