OpenClaw Telegram ACP Troubleshooter

• Outcome: Pass
• Latest audit: Sep 12, 2026 · 6:49 AM UTC
• Version: 1.0.1

Overview

This is a troubleshooting-only OpenClaw Telegram skill whose log and session checks fit its stated purpose, with some privacy and operational cautions for users.

Before installing or using this skill, treat OpenClaw command logs, codex session files, bot identifiers, tokens, group IDs, and message bodies as sensitive. Inspect only the relevant time window and topic/session, redact raw values before sharing diagnostics, and confirm before taking service-impacting actions like removing the bot, promoting it to admin, or restarting the gateway.

A.I.G

Vulnerability Patterns

• Skill Instruction Hijacking: Alters the agent's session goals or safety constraints when the skill loads
• Agent Memory Poisoning: Writes attacker-controlled rules into memory that affect later sessions
• Remote Payload Retrieval and Execution: Fetches external code whose behavior can change after review
• Embedded Malicious Code: Ships malicious scripts inside the skill and executes them locally
• Unauthorized Access and Privilege Escalation: Obtains permissions beyond the task's legitimate needs

Findings (1)

other

Note

• Location: SKILL.md:43

Finding

Potential Exposure of Sensitive Agent Session and Message Data

Content

View full analysis

Vulnerability Details

File Locations:

• SKILL.md:43-46
• SKILL.md:78-80
• TROUBLESHOOTING.md:68-90
• FAQ.md:30-31

Vulnerability Type: other: Sensitive Agent Session Data Access
Risk Level: Low

Relevant Instructions

SKILL.md:43-46:

markdown

3. Confirm OpenClaw ingestion:
   - watch `~/.openclaw/logs/commands.log`
   - inspect the relevant session file under `~/.openclaw/agents/codex/sessions/`

SKILL.md:78-80:

bash

systemctl --user show -p MainPID -p ActiveEnterTimestamp openclaw-gateway
tail -f ~/.openclaw/logs/commands.log
tail -f /tmp/openclaw/openclaw-$(date +%F).log

TROUBLESHOOTING.md:68-90:

markdown

### Command Log

```bash
tail -f ~/.openclaw/logs/commands.log

Expected topic session key shape:

text

agent:codex:telegram:group:<group_id>:topic:<topic_id>

Session Proof

Inspect the matching file under:

text

~/.openclaw/agents/codex/sessions/

If /new appears there but later plain text does not, the issue is upstream of ACP.

text

`FAQ.md:30-31`:

```markdown
Check `~/.openclaw/logs/commands.log` and the matching file in `~/.openclaw/agents/codex/sessions/`.

Technical Analysis

The skill instructs the executing agent or operator to inspect OpenClaw command logs and Codex session files. These resources can contain user message content, conversation history, group or topic identifiers, session identifiers, and other operational metadata.

This access is related to the stated troubleshooting purpose and is not evidence of malicious intent. The skill also instructs users not to reveal secret values. However, it does not require explicit approval before reading session records, define a minimum necessary scope, or require redaction of message bodies and identifiers before diagnostic evidence is returned.

Consequently, sensitive local conversation data could enter the age ...[truncated 1656 chars]

Remediation

View remediation

Remediation Suggestions

1. Require explicit user approval before opening command logs or session files that may contain conversation content.
2. Begin with metadata-only checks, such as file existence, modification timestamps, session-key presence, and narrowly filtered event counts.
3. Restrict reads to the specific topic, session, and time window relevant to the reported incident.
4. Avoid recursively listing or reading the entire Codex session directory.
5. Redact message bodies, tokens, usernames, group identifiers, topic identifiers, and session identifiers before including evidence in a response.
6. Add an instruction that raw logs and complete session records must never be reproduced in diagnostic output.
7. Prefer bounded commands over unrestricted live output. For example, use narrowly filtered and line-limited reads rather than an open-ended tail -f.
8. Document appropriate filesystem permissions so session files and logs are readable only by the OpenClaw service account and authorized operators.
9. Preserve the existing prohibition against revealing secrets and expand it to cover personal messages, conversation history, and routing metadata.

SkillSpector

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (2)

Missing User Warnings

Medium

• Category: Not specified by scanner
• Confidence: 85% confidence

Finding

This is a markdown file, so SQP-2 applies to missing warnings in the skill description. The guide instructs operators to verify the active token and inspect command/session logs, which can expose bot credentials and stored user content, but it provides no caution about handling sensitive data safely.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low

• Category: Not specified by scanner
• Confidence: 77% confidence

Finding

The markdown recommends removing and re-adding the bot, then promoting it to admin, which can temporarily interrupt automation and affect group configuration. Because the document does not warn users about this operational impact, it omits a useful safety disclosure for an action affecting system integrity and availability.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.