Back to skill

Security audit

Huo15 Multi Agent

Security checks for vulnerabilities and agentic risk

Overview

This skill is a multi-agent helper, but its bundled scripts handle task and worker inputs unsafely enough to risk local file changes or command execution if given crafted values.

Review before installing. Do not run the bundled scripts with untrusted task names, worker IDs, or results, and avoid enabling exec/process for subagents until the author adds strict input validation, safe JSON serialization, confirmation for deletion, and clear permission limits.

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 (3)

T09 · Insecure Skill Coding Practices

Error
Location
scripts/coordinator.sh:106
Finding

Arbitrary Python Code Execution Through Shell Argument Interpolation

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Error
Location
scripts/coordinator.sh:89
Finding

Path Traversal Through Unvalidated Task and Worker Identifiers

Content
View full analysis
"$TEAM_DIR/tasks/$task_id.json" << EOF { "taskId": "$task_id", "description": "$task_desc", "status": "pending", "assignedTo": null, "result": null, "createdAt": "$(date -u +%Y-%m-%dT%H:%M:%SZ)", "completedAt": null } EOF ``` From `scripts/team.sh:54-65`: ```bash # 创建 worker 配置 mkdir -p "$TEAM_DIR/workers/$worker_id" cat > "$TEAM_DIR/workers/$worker_id/worker.json" << EOF { "workerId": "$worker_id", "task": "$task_desc", "status": "running", "startedAt": "$(date -u +%Y-%m-%dT%H:%M:%SZ)", "result": null } EOF ``` From `scripts/team.sh:124-130`: ```bash local worker_file="$TEAM_DIR/workers/$worker_id/worker.json" if [ ! -f "$worker_file" ]; then echo "❌ Worker 不存在: $worker_id" return 1 fi ``` ### Technical Analysis `task_id` and `worker_id` are inserted directly into filesystem paths without validating that they are simple identifiers. Quoting a shell path prevents word splitting and wildcard expansion, but it does not prevent the operating system from resolving `..` components. Consequently, an identifier containing sequences such as `../../` can escape the intended `tasks` or `workers` directory. The scripts can then: - Truncate and rewrite a resolved `.json` file. - Create an attacker-selected directory path ending in a generated `worker.json`. - Read or update a `worker.json` outside the intended worker directory. - Follow pre-existing symbolic links because no symlink checks are performed. The coordinator path always appends `.json`, while the worker path always appends `/worker.json`; these constraints reduce the choice of target filename but do not enforce containment wit ...[truncated 1718 chars]
Remediation
View remediation
&2 return 1 ;; esac [ "${#1}" -le 64 ] || { echo "Identifier is too long" >&2 return 1 } } ``` 2. Call the validation function before every filesystem or Python operation involving an identifier. 3. Canonicalize the parent directory and candidate path, then verify that the candidate remains beneath the expected root. 4. Reject identifiers containing `/`, `\`, `.` path components, control characters, or leading hyphens. 5. Reject symbolic links in writable path components and destination files. 6. Create files atomically with restrictive permissions, such as a temporary file under the validated directory followed by `mv`. 7. Set a restrictive `umask`, for example `umask 077`, before creating state files. 8. Add tests covering `../`, repeated traversal, symlink destinations, long identifiers, control characters, and Unicode path edge cases. ]]>

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/coordinator.sh:53
Finding

JSON Injection and State Corruption Through Manual JSON Construction

Content
View full analysis
"$TEAM_DIR/config.json" << EOF { "teamName": "$TEAM_NAME", "coordinator": "$COORDINATOR", "startedAt": "$(date -u +%Y-%m-%dT%H:%M:%SZ)", "status": "active", "tasks": [] } EOF ``` From `scripts/coordinator.sh:91-101`: ```bash cat > "$TEAM_DIR/tasks/$task_id.json" << EOF { "taskId": "$task_id", "description": "$task_desc", "status": "pending", "assignedTo": null, "result": null, "createdAt": "$(date -u +%Y-%m-%dT%H:%M:%SZ)", "completedAt": null } EOF ``` From `scripts/team.sh:29-36`: ```bash cat > "$TEAM_DIR/team.json" << EOF { "teamName": "$team_name", "createdAt": "$(date -u +%Y-%m-%dT%H:%M:%SZ)", "members": [], "status": "active" } EOF ``` From `scripts/team.sh:55-65`: ```bash cat > "$TEAM_DIR/workers/$worker_id/worker.json" << EOF { "workerId": "$worker_id", "task": "$task_desc", "status": "running", "startedAt": "$(date -u +%Y-%m-%dT%H:%M:%SZ)", "result": null } EOF ``` ### Technical Analysis The scripts manually build JSON documents by inserting variables into heredoc templates. JSON requires quotation marks, backslashes, newlines, tabs, control characters, and other special values to be escaped according to the JSON grammar. The scripts perform no such encoding. A command-line value containing a quotation mark, backslash, or newline can therefore: - Produce syntactically invalid JSON. - Terminate an intended JSON string. - Inject additional properties or values. - Change how downstream readers interpret the state file. This issue is distinct from Python source injection: it affects files constructed directly by shell heredocs, even if no Python process subsequently exec ...[truncated 1380 chars]
Remediation
View remediation
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
Findings (16)

Tp4

High
Category
MCP Tool Poisoning
Confidence
91% confidence
Finding

The description presents this skill as a functioning multi-agent parallel work system with coordinator behavior, task distribution, and result summarization based on OpenClaw sessions_spawn. The supplied code does not implement those core behaviors. It only validates arguments, creates a log directory, writes a simple log entry, generates a default label, and prints explanatory text claiming OpenClaw will do the rest. There is no actual sessions_spawn call, no subagent creation, no coordinator logic, no task routing, and no result collection. This is a material description-behavior mismatch because the primary advertised capability is absent from the implementation.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
94% confidence
Finding

The description promises a multi-agent parallel work system built on OpenClaw sessions_spawn with coordinator mode, task assignment, and result summarization. The code only provides lightweight local team bookkeeping in shell/Python: it creates team.json and worker.json files, updates status fields, lists them, and deletes them. There is no invocation of OpenClaw session spawning, no process/thread management, no coordinator logic, and no true parallel execution. The code’s primary purpose is team/worker state tracking rather than an operational multi-agent collaboration engine. There is also a functional bug where the main dispatcher calls leave_worker, but the defined function is leave_team.

Content

No source excerpt is available for this finding.

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
95% confidence
Finding

rm -rf "$TEAM_DIR/workers/$worker_id" uses a path derived from untrusted input without canonicalization or strict validation. An attacker could supply traversal sequences such as ../../... to delete files or directories outside the intended workers area, leading to destructive arbitrary file deletion within the user's permissions.

Content

Scanner excerpt · scripts/team.sh (reported line 224)May include surrounding context.

sh
return 1
    fi
    
    rm -rf "$TEAM_DIR/workers/$worker_id"
    
    # 从团队成员中移除
    python3 << PYEOF

Missing User Warnings

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

Bulk cleanup removes the entire workers directory recursively without any confirmation or backup, causing immediate loss of all worker state. In this skill's coordination context, that can wipe active task tracking and results for all agents, creating a high availability and integrity impact.

Content

No source excerpt is available for this finding.

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
95% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · scripts/team.sh (reported line 266)May include surrounding context.

sh
done
    
    # 清理目录
    rm -rf "$TEAM_DIR/workers"
    mkdir -p "$TEAM_DIR/workers"
    
    echo "✅ 清理完成"

Undeclared Tool Scope

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.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The trigger example uses broad natural-language phrasing like '帮我同时处理...' that could overlap with ordinary user requests, increasing the chance of accidental invocation. In a skill that discusses spawning multiple workers and parallel execution, unintended activation can amplify actions, consume resources, or cause task execution the user did not explicitly intend.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The documentation grants subagents potentially dangerous capabilities such as 'exec' and 'process' but does not warn users about command-execution, lateral action, or resource-abuse risks. In a multi-agent context, these permissions are more dangerous because tasks can be parallelized and delegated, increasing the blast radius of mistakes or malicious prompts.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

This shell script is a code file, so SQP-3 applies to its natural-language strings. The example invocation and all demonstrated agent commands are presented only in Chinese, effectively prescribing a specific language for user interaction without any opt-in or explanation that the skill is region-specific.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The script persists the full task string to a markdown log under the user's home directory without any minimization, redaction, consent prompt, or retention controls. In a multi-agent workflow, tasks may contain sensitive prompts, source snippets, credentials, incident details, or proprietary data, so this creates a realistic confidentiality risk if the log is later read by other tools, users, backups, or synced storage.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

User-facing comments, status messages, and help text are consistently presented in Chinese, but the script does not offer any language selection or explain that it is intended only for a Chinese-speaking context. This creates a language-policy issue because the skill effectively forces a specific language without user opt-in.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The header and function comments state that the script can start work agents and, during cleanup, stop running workers. In practice, startup only writes worker.json metadata, and cleanup merely prints a message with a comment acknowledging that a stop signal should be sent, so the documentation overstates the implemented behavior.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The manifest describes a multi-agent parallel work system based on OpenClaw session spawning, coordinator mode, task assignment, and result aggregation. In this file, the advertised worker startup path only creates directories and JSON records, with no call to any agent runtime, session spawn mechanism, or coordinator logic, so the implemented behavior is a bookkeeping stub rather than the claimed multi-agent execution capability.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
85% confidence
Finding

The script includes destructive state-removal operations that exceed simple coordination and are exposed without safeguards. In a multi-agent skill context, deleting worker state can cause denial of service, loss of audit/history data, and accidental destruction of unrelated files if path inputs are abused.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The script performs irreversible deletion of worker data via rm -rf without any confirmation, dry-run, or warning. Because the deletion target incorporates a user-supplied worker_id, accidental or malicious path manipulation could remove more data than intended and disrupt the system.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
91% confidence
Finding

Natural-language strings in comments, usage examples, and runtime output are presented only in Chinese, which can violate language/locale policy when no user opt-in or justification is provided. The file does not indicate that the skill is intentionally limited to a Chinese-speaking audience or region-specific environment.

Content

No source excerpt is available for this finding.