Back to skill

Security audit

newsbot-digest

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this skill suspicious with medium confidence, mainly because a reported browser redirect risk could send requests to private network services; install only if you can accept that risk and review the incomplete evidence.

The verdict is suspicious, so consider the reported private-network redirect risk before installing, especially in an environment that can reach sensitive internal services. Also note that dependencies are unpinned and the scan is marked partial because some local references could not be resolved; the evidence does not establish that listed dependencies are vulnerable or that the skill is malicious.

Vulnerability Patterns
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Findings (1)

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/newsbot/articles.py
Finding

Automatically selected publisher pages can redirect the browser into private networks

Content
View full analysis

Impact

The attacker can cause requests to internal or loopback services, including GET endpoints with side effects. If the destination returns extractable internal text, that text is sent to the configured external LLM provider during validation. Impact depends on services reachable from the execution environment; credential theft or code execution is not established.

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Rogue AgentSelf-Modification, Session Persistence
Findings (49)

Credential Access

High
Category
Privilege Escalation
Confidence
60% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · scripts/newsbot/config.py (reported line 111)May include surrounding context.

python
# ---------------- окружение и логгер ----------------
def load_env(extra_env: Optional[Path], workdir: Path) -> None:
    for p in (extra_env, workdir / ".env", SKILL_DIR / ".env", Path.cwd() / ".env"):
        if p and Path(p).is_file():
            load_dotenv(p, override=False)

Credential Access

High
Category
Privilege Escalation
Confidence
60% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · scripts/newsbot/config.py (reported line 111)May include surrounding context.

python
# ---------------- окружение и логгер ----------------
def load_env(extra_env: Optional[Path], workdir: Path) -> None:
    for p in (extra_env, workdir / ".env", SKILL_DIR / ".env", Path.cwd() / ".env"):
        if p and Path(p).is_file():
            load_dotenv(p, override=False)

Credential Access

High
Category
Privilege Escalation
Confidence
60% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · scripts/newsbot/config.py (reported line 111)May include surrounding context.

python
# ---------------- окружение и логгер ----------------
def load_env(extra_env: Optional[Path], workdir: Path) -> None:
    for p in (extra_env, workdir / ".env", SKILL_DIR / ".env", Path.cwd() / ".env"):
        if p and Path(p).is_file():
            load_dotenv(p, override=False)

Credential Access

High
Category
Privilege Escalation
Confidence
60% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · scripts/run_digest.py (reported line 31)May include surrounding context.

python
help="Папка состояния (история выборок для дедупликации). Должна сохраняться между запусками.")
    p.add_argument("--output-dir", default=os.getenv("NEWSBOT_OUTPUT_DIR"),
                   help="Куда класть дайджест и PDF (по умолчанию <workdir>/outputs)")
    p.add_argument("--env", default=None, help="Путь к .env с ключами")
    p.add_argument("--browser", choices=["headless", "virtual"], default=os.getenv("NEWSBOT_BROWSER", "headless"),
                   help="headless — без окна (по умолчанию); virtual — через Xvfb, тоже без окна")
    p.add_argument("--skip-telegram", action="store_true")

Ae4

Medium
Category
analysis-evasion
Confidence
80% confidence
Finding

Suspicious Unicode normalization or mixed-script content

Content

No source excerpt is available for this finding.

Skill declares no tool scope ('permissions' or 'allowed-tools') but code capabilities were detected: env, file_read, file_write, network.

Medium
Category
MCP Least Privilege
Confidence
70% confidence
Finding

Without declared permissions the skill's intent is opaque and cannot be validated.

Content

No source excerpt is available for this finding.

Session Persistence

Medium
Category
Rogue Agent
Confidence
65% confidence
Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · SKILL.md (reported line 49)May include surrounding context.

bash
cd "$SKILL_DIR"
nohup python scripts/run_digest.py --workdir "$WORKDIR" --output-dir "$OUTDIR" > "$WORKDIR/run.log" 2>&1 &

Затем раз в 2–5 минут проверяй "$OUTDIR/$(date +%F)/status.json": поле current показывает текущий этап,

The prompt mandates Russian output without an explicit user language choice or opt-in.

Medium
Category
Output Handling
Confidence
94% confidence
Finding

Line 1 explicitly requires translating the article into Russian, while lines 9 and 13 reinforce Russian-language phrasing and formatting. The file provides no user-selectable target language or explicit opt-in to this language constraint.

Content

No source excerpt is available for this finding.

Ae4

Medium
Category
analysis-evasion
Confidence
80% confidence
Finding

Suspicious Unicode normalization or mixed-script content

Content

No source excerpt is available for this finding.

Ae4

Medium
Category
analysis-evasion
Confidence
80% confidence
Finding

Suspicious Unicode normalization or mixed-script content

Content

No source excerpt is available for this finding.

Ae4

Medium
Category
analysis-evasion
Confidence
80% confidence
Finding

Suspicious Unicode normalization or mixed-script content

Content

No source excerpt is available for this finding.

Ae4

Medium
Category
analysis-evasion
Confidence
80% confidence
Finding

Suspicious Unicode normalization or mixed-script content

Content

No source excerpt is available for this finding.

The prompt requires all responses to be in Russian without user opt-in.

Medium
Category
Output Handling
Confidence
99% confidence
Finding

The instruction “Ответ всегда пиши на русском языке” explicitly requires responses always to be written in Russian, including when the input article is in English. The file offers no language choice or justification for a fixed-language requirement.

Content

No source excerpt is available for this finding.

Ae4

Medium
Category
analysis-evasion
Confidence
80% confidence
Finding

Suspicious Unicode normalization or mixed-script content

Content

No source excerpt is available for this finding.

Ae4

Medium
Category
analysis-evasion
Confidence
80% confidence
Finding

Suspicious Unicode normalization or mixed-script content

Content

No source excerpt is available for this finding.

Ae4

Medium
Category
analysis-evasion
Confidence
80% confidence
Finding

Suspicious Unicode normalization or mixed-script content

Content

No source excerpt is available for this finding.

Ae4

Medium
Category
analysis-evasion
Confidence
80% confidence
Finding

Suspicious Unicode normalization or mixed-script content

Content

No source excerpt is available for this finding.

Ae4

Medium
Category
analysis-evasion
Confidence
80% confidence
Finding

Suspicious Unicode normalization or mixed-script content

Content

No source excerpt is available for this finding.

Ae4

Medium
Category
analysis-evasion
Confidence
80% confidence
Finding

Suspicious Unicode normalization or mixed-script content

Content

No source excerpt is available for this finding.

The workflow mandates Russian translations and summaries without offering a language choice.

Low
Category
Output Handling
Confidence
92% confidence
Finding

Steps 5 and 6 explicitly require article translations and headline summaries in Russian. Although this output language is documented, the skill neither asks the user to opt in nor explains a locale-specific requirement that justifies enforcing it.

Content

No source excerpt is available for this finding.

The selection prompt instructs the agent to select 10 news items despite the manifest promising a top-5 digest.

Low
Category
Excessive Agency
Confidence
99% confidence
Finding

L079 explicitly requests a final list of 10 IDs, and L223 repeats the instruction to select a top 10. This conflicts with the manifest's top-5 scope and the five-item output instructions at L203–L210, leaving the agent with contradictory requirements for the final selection size.

Content

No source excerpt is available for this finding.

The selection prompt requires ten news items, while the manifest advertises a top-five digest.

Low
Category
Excessive Agency
Confidence
90% confidence
Finding

The manifest describes a daily top-five digest, but L046 explicitly requires a final list of ten IDs. The ten-item response template at L174 and repeated instruction at L190 reinforce this different selection contract; whether another component later reduces the selection to five is not shown.

Content

No source excerpt is available for this finding.

Unpinned Dependencies

Low
Category
Supply Chain
Confidence
99% confidence
Finding

The pandas dependency has no version constraint, so installations can resolve to different, unreviewed releases. This is a dependency-management weakness, not evidence that an installed version is vulnerable or malicious.

Content

Scanner excerpt · requirements.txt (reported line 1)May include surrounding context.

text
pandas
openpyxl
requests
beautifulsoup4

Unverifiable Dependency: pandas has 1 known advisory(ies) (CVE-2020-13091 (** DISPUTED ** pandas through 1.0.3 can unserialize and execute commands from an)), but the manifest does not pin a version, so it is unknown whether the installed release is affected

Low
Category
Supply Chain
Confidence
40% confidence
Finding

OSV returned advisories for this package, but its resolved version is unknown. A package-name lookup does not establish whether the installed release is affected by those advisories.

Content

No source excerpt is available for this finding.

Unpinned Dependencies

Low
Category
Supply Chain
Confidence
99% confidence
Finding

The openpyxl dependency is unconstrained, allowing installations to drift between releases without review. Spreadsheet parsing can expose an application to hostile inputs, but this manifest does not establish that the skill processes untrusted spreadsheets.

Content

Scanner excerpt · requirements.txt (reported line 2)May include surrounding context.

text
pandas
openpyxl
requests
beautifulsoup4
lxml