Back to skill

Security audit

Session Recover

Security checks for vulnerabilities and agentic risk

Overview

This skill is a session recovery tool, but it is too broad because it can expose arbitrary archived conversations, hidden thinking fields, and partial tool results without clear authorization or privacy controls.

Install only if you are comfortable giving the agent a workflow for reading local OpenClaw session archive files. Before use, restrict it to sessions you own or are explicitly authorized to inspect, avoid parsing arbitrary private/group/reset archives, and do not expose thinking fields, tool results, credentials, tokens, or sensitive code in recovery reports.

Vulnerability Patterns
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
  • 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)

T05 · Unauthorized Access and Privilege Escalation

Warning
Location
references/parse_session.py:32
Finding

Unrestricted Session Archive Parsing Exposes Private Conversation Data

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Findings (10)

Ssd 3

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

The skill instructs recovery and summarization of arbitrary session transcripts, including private conversation content, which creates a direct confidentiality risk. In the context of an agent environment, session archives can contain credentials, personal data, proprietary code, and sensitive operational history far beyond what a recall helper should expose.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

High
Category
Not specified by scanner
Confidence
99% confidence
Finding

The skill explicitly instructs extraction of message.content[].thinking, which is hidden internal reasoning rather than normal user-visible transcript content. Exposing or summarizing internal reasoning can leak confidential intermediate analysis, policy logic, or secrets that should never be surfaced through a recall feature.

Content

No source excerpt is available for this finding.

Ssd 3

High
Category
Not specified by scanner
Confidence
99% confidence
Finding

Directing extraction of both message text and thinking fields turns the skill into a mechanism for disclosing internal and private data from archives. This is especially dangerous because thinking fields may contain hidden model reasoning or other non-user-facing content that should remain inaccessible even when transcript recovery is allowed.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
96% confidence
Finding

The skill clearly instructs reading local session archive files and parsing transcript data, yet the manifest declares no explicit tool scope or permission boundary. This makes the skill's data-access behavior implicit and increases the risk of overbroad file access without informed review or consent.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill advertises full session recovery and transcript parsing without warning that it may reveal sensitive private history, secrets, or personal data. Users may invoke it expecting benign summarization, while the actual behavior materially increases exposure of archived conversations.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The description expands from recovering the current or previous session to letting any user retrieve the complete transcript of arbitrary sessions. That broadens the skill from continuity support into general transcript exfiltration, which can expose unrelated private conversations and sensitive historical context.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The documented commands enumerate and read local archived and reset session files from disk without disclosing the sensitivity of those artifacts. Reset archives and local session stores may contain data users assumed was lost or private, so undisclosed access materially increases the chance of unintended disclosure.

Content

No source excerpt is available for this finding.

Ssd 3

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The output template encourages reproducing key context, configuration values, code snippets, and recent instructions from recovered sessions, which can amplify leakage of secrets and sensitive implementation details. Even if the recovery target is valid, this presentation increases downstream exposure by copying high-value data into a new response surface.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
76% confidence
Finding

The natural-language instructions, usage cues, and trigger description are presented in Chinese, which may effectively force a specific language for use of the skill. The file does not offer an opt-in language choice or explain that the skill is intentionally limited to a Chinese-language context.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
95% confidence
Finding

This code file contains natural-language strings for the description, usage text, and help output entirely in Chinese. Under the policy rule for language/locale, forcing a specific language without user opt-in is a natural-language policy concern when no alternative or justification is provided.

Content

No source excerpt is available for this finding.