fix(update): reject symlink members in update ZIP

_update_via_zip downloads a source ZIP from GitHub and calls
zipfile.ZipFile.extractall. The existing zip-slip path guard validates
each member's path stays under tmp_dir, but does not check member type
— so a ZIP containing a symlink member would still be materialized by
extractall, and a symlink target could point outside the extracted
tree (or to a sensitive system path).

This isn't a high-likelihood threat for hermes-agent's actual GitHub
source ZIPs (we don't ship symlinks), but the extractall path runs as
the user's account and a compromised mirror could plant arbitrary files
via the symlink → target → write chain.

Reject any member whose Unix mode bits (upper 16 bits of external_attr)
are S_IFLNK before extractall. Hermes source ZIPs contain only regular
files and directories; a symlink member is unambiguously suspicious.

Regression tests cover: symlink member rejection (raises ValueError,
caught by the outer try/except as a clean SystemExit, no extraction),
and the happy-path verification that a normal ZIP doesn't trigger the
symlink reject message.

Salvaged from PR #15881 by @codeblackhole1024. The remaining pieces of
that PR were already on main or contradicted explicit design decisions:
- config.yaml write-deny: already in agent/file_safety.py's
  control_file_names denylist (the modern guard); the proposed addition
  to build_write_denied_paths was the legacy path.
- Quick commands danger detection: contradicts the explicit
  cli.py:8491-8492 comment 'shell=True is intentional: quick_commands
  are user-defined shell snippets from config.yaml — not agent/LLM
  controlled.'
- Memory plugin shlex.split for dep checks: already on main
  (hermes_cli/memory_setup.py:133).

Co-authored-by: teknium1 <127238744+teknium1@users.noreply.github.com>
This commit is contained in:
codeblackhole1024 2026-05-25 05:15:19 -07:00 committed by Teknium
parent 5f20322d23
commit bd2756dd22
2 changed files with 129 additions and 1 deletions

View file

@ -7000,8 +7000,13 @@ def _update_via_zip(args):
urlretrieve(zip_url, zip_path)
print("→ Extracting...")
import stat as _stat
with zipfile.ZipFile(zip_path, "r") as zf:
# Validate paths to prevent zip-slip (path traversal)
# Validate paths to prevent zip-slip (path traversal) AND reject
# symlink members. A GitHub source ZIP for hermes-agent itself
# should never contain symlinks — they'd point outside the
# extracted tree and let an attacker who can compromise the
# update mirror plant arbitrary files via the update path.
tmp_dir_real = os.path.realpath(tmp_dir)
for member in zf.infolist():
member_path = os.path.realpath(os.path.join(tmp_dir, member.filename))
@ -7012,6 +7017,13 @@ def _update_via_zip(args):
raise ValueError(
f"Zip-slip detected: {member.filename} escapes extraction directory"
)
# Unix mode lives in the upper 16 bits of external_attr;
# mask to the file-type bits.
mode = (member.external_attr >> 16) & 0o170000
if _stat.S_ISLNK(mode):
raise ValueError(
f"ZIP contains unsupported symlink member: {member.filename}"
)
zf.extractall(tmp_dir)
# GitHub ZIPs extract to hermes-agent-<branch>/