cleveragents/cleveragents-core

Fork 3

BUG-HUNT: [boundary] security_scanner.py flags legitimate subprocess.run usage in Python source files causing false positive security alerts #7387

New issue

Open

opened 2026-04-10 18:46:21 +00:00 by HAL9000 · 3 comments

HAL9000 commented

2026-04-10 18:46:21 +00:00

Owner

Bug Report: [boundary] ConfigSecurityScanner incorrectly scans Python/code files, producing false positives for legitimate subprocess usage

Severity Assessment

Impact: The scan_file() function scans any file regardless of its type. When applied to Python source files (.py), shell scripts, or other code files, it will flag every legitimate use of subprocess.run, eval(), os.system(), etc. as security violations. This makes the scanner useless for mixed-file scanning and can cause CI failures on legitimate code.
Likelihood: High — the scanner is designed for YAML/TOML config files but accepts any path. Any automated pipeline that points it at a code directory will get massive false positives.
Priority: Medium

Location

File: src/cleveragents/config/security_scanner.py
Function/Class: scan_file(), scan_content()
Lines: ~130-155

Description

The security scanner's docstring says it is designed to scan "YAML, TOML, and generic config files" for code injection vulnerabilities. However, the implementation has no file type checking — it will scan any file, including Python source files.

The patterns it scans for (eval(, exec(, subprocess.run, etc.) are legitimate and necessary in Python source code. When the scanner is applied to any Python file, it will generate false positive violations.

The _register("subprocess.run", ...) pattern is particularly problematic because:

The security_scanner.py file itself uses scan_content() to scan files
If someone scans the security_scanner.py source file, the pattern registration lines themselves might trigger false positives
More importantly, the git_tools.py file in this codebase contains subprocess.run() calls which ARE legitimate

Additionally, there's a subtle issue in the _strip_inline_comment() function: it only handles YAML/TOML comment syntax (#), but it also handles ; (used in INI files). For Python files, # is also the comment character, so Python comment stripping would work. However, multi-line Python strings containing the patterns would still be flagged.

Evidence

# From security_scanner.py - registers these as violation patterns:
_register("subprocess.run", Severity.CRITICAL, "subprocess.run() executes shell commands")
_register("eval(", Severity.CRITICAL, "eval() can execute arbitrary Python code")

# From tool/builtins/git_tools.py - legitimate usage in code files:
result = subprocess.run(
    cmd,
    cwd=cwd,
    capture_output=True,
    text=True,
    check=True,
    timeout=timeout,
    env=env,
)

# If scan_file("src/cleveragents/tool/builtins/git_tools.py") were called:
# -> 5 CRITICAL violations for every subprocess.run() in the file!
# This is a false positive — subprocess.run() is fine in Python code

Expected Behavior

The scanner should either:

Only accept YAML/TOML/INI config files (validate extension before scanning)
Have a mode flag --code-files vs --config-files
Document clearly that it's NOT for Python source files

Actual Behavior

The scanner accepts any file path and scans for patterns that are legitimate in code files. This will produce false positives for any Python file containing subprocess calls or eval usage.

TDD Note

After this bug issue is verified, a corresponding Type/Testing issue will be created for TDD. The test will use tags: @tdd_issue, @tdd_issue_, and @tdd_expected_fail to prove the bug exists before fixing it.

Automated by CleverAgents Bot
Supervisor: Bug Detection Pool | Agent: bug-hunt-pool-supervisor

## Bug Report: [boundary] ConfigSecurityScanner incorrectly scans Python/code files, producing false positives for legitimate subprocess usage ### Severity Assessment - **Impact**: The `scan_file()` function scans any file regardless of its type. When applied to Python source files (`.py`), shell scripts, or other code files, it will flag every legitimate use of `subprocess.run`, `eval()`, `os.system()`, etc. as security violations. This makes the scanner useless for mixed-file scanning and can cause CI failures on legitimate code. - **Likelihood**: High — the scanner is designed for YAML/TOML config files but accepts any path. Any automated pipeline that points it at a code directory will get massive false positives. - **Priority**: Medium ### Location - **File**: `src/cleveragents/config/security_scanner.py` - **Function/Class**: `scan_file()`, `scan_content()` - **Lines**: ~130-155 ### Description The security scanner's docstring says it is designed to scan "YAML, TOML, and generic config files" for code injection vulnerabilities. However, the implementation has no file type checking — it will scan any file, including Python source files. The patterns it scans for (`eval(`, `exec(`, `subprocess.run`, etc.) are **legitimate and necessary** in Python source code. When the scanner is applied to any Python file, it will generate false positive violations. The `_register("subprocess.run", ...)` pattern is particularly problematic because: 1. The `security_scanner.py` file itself uses `scan_content()` to scan files 2. If someone scans the `security_scanner.py` source file, the pattern registration lines themselves might trigger false positives 3. More importantly, the git_tools.py file in this codebase contains `subprocess.run()` calls which ARE legitimate Additionally, there's a subtle issue in the `_strip_inline_comment()` function: it only handles YAML/TOML comment syntax (`#`), but it also handles `;` (used in INI files). For Python files, `#` is also the comment character, so Python comment stripping would work. However, multi-line Python strings containing the patterns would still be flagged. ### Evidence ```python # From security_scanner.py - registers these as violation patterns: _register("subprocess.run", Severity.CRITICAL, "subprocess.run() executes shell commands") _register("eval(", Severity.CRITICAL, "eval() can execute arbitrary Python code") # From tool/builtins/git_tools.py - legitimate usage in code files: result = subprocess.run( cmd, cwd=cwd, capture_output=True, text=True, check=True, timeout=timeout, env=env, ) # If scan_file("src/cleveragents/tool/builtins/git_tools.py") were called: # -> 5 CRITICAL violations for every subprocess.run() in the file! # This is a false positive — subprocess.run() is fine in Python code ``` ### Expected Behavior The scanner should either: 1. Only accept YAML/TOML/INI config files (validate extension before scanning) 2. Have a mode flag `--code-files` vs `--config-files` 3. Document clearly that it's NOT for Python source files ### Actual Behavior The scanner accepts any file path and scans for patterns that are legitimate in code files. This will produce false positives for any Python file containing subprocess calls or eval usage. ### Category boundary ### TDD Note After this bug issue is verified, a corresponding Type/Testing issue will be created for TDD. The test will use tags: @tdd_issue, @tdd_issue_<this-issue-number>, and @tdd_expected_fail to prove the bug exists before fixing it. --- **Automated by CleverAgents Bot** Supervisor: Bug Detection Pool | Agent: bug-hunt-pool-supervisor

HAL9000 referenced this issue

2026-04-10 18:51:41 +00:00

[AUTO-BUG-SUP] Bug Hunt Status (Cycle 4) #7384

HAL9000 referenced this issue

2026-04-10 19:06:48 +00:00

[AUTO-ARCH] Architecture Status (Cycle 1) #7343

HAL9000 added the

State

Unverified

label

2026-04-10 19:35:10 +00:00

freemo added the

labels

2026-04-12 18:35:32 +00:00

HAL9000 added

and removed

labels

2026-04-14 06:24:07 +00:00

HAL9000 commented

2026-04-14 06:24:07 +00:00

Author

Owner

✅ Verified — Bug: security scanner false positives for legitimate subprocess.run in Python files. MoSCoW: Should-have. Priority: Medium.

Automated by CleverAgents Bot
Supervisor: Project Owner | Agent: project-owner-pool-supervisor

✅ **Verified** — Bug: security scanner false positives for legitimate subprocess.run in Python files. MoSCoW: Should-have. Priority: Medium. --- **Automated by CleverAgents Bot** Supervisor: Project Owner | Agent: project-owner-pool-supervisor

HAL9000 commented

2026-04-14 06:29:45 +00:00

Author

Owner

✅ Verified — Bug: security scanner false positives for legitimate subprocess.run in Python files. MoSCoW: Should-have. Priority: Medium.

Automated by CleverAgents Bot
Supervisor: Project Owner | Agent: project-owner-pool-supervisor