Back to skill

Security audit

Session Archive

Security checks for vulnerabilities and agentic risk

Overview

This plugin does what it claims, but it automatically keeps a long-lived local archive of full conversations and operational metadata without clear opt-in, retention, encryption, or scope controls.

Install only if you intentionally want all OpenClaw conversations and operational records archived locally. Treat the database and its WAL/SHM files as sensitive, choose a private path, restrict file permissions, avoid storing secrets in archived sessions, and confirm you have a deletion or retention process before using it broadly.

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
src/db.ts:68
Finding

Unencrypted Persistent Storage of Sensitive Conversation and Operation Data

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Findings (32)

Known Vulnerable Dependency: @mariozechner/pi-coding-agent==0.63.1 — 3 advisory(ies): CVE-2026-54326 (Pi Agent: Potential XSS in HTML session exports via Markdown URL sanitization by); CVE-2026-54328 (Pi Agent: Predictable temporary extension install paths allow local privilege es); CVE-2026-54327 (Pi Agent: Race condition in Pi auth.json writes could expose stored credentials)

High
Category
Supply Chain
Confidence
97% confidence
Finding

The presence of @mariozechner/pi-coding-agent 0.63.1 is a meaningful risk because the advisories include credential exposure, predictable temporary paths, and XSS in exported HTML. In this skill context, an agent-oriented package with filesystem and session export capabilities increases the chance that vulnerable features are reachable and could affect local secrets or generated artifacts.

Content

No source excerpt is available for this finding.

Known Vulnerable Dependency: axios==1.14.0 — 16 advisory(ies): CVE-2026-44494 (axios Vulnerable to Full Man-in-the-Middle via Prototype Pollution Gadget in `co); CVE-2026-44495 (axios Vulnerable to Credential Theft and Response Hijacking via Prototype Pollut); CVE-2025-62718 (Axios has a NO_PROXY Hostname Normalization Bypass that Leads to SSRF) +13 more

High
Category
Supply Chain
Confidence
96% confidence
Finding

axios 1.14.0 is listed with numerous advisories including SSRF-related NO_PROXY bypass and prototype-pollution-linked request manipulation issues. In an agent ecosystem that may fetch remote resources, call APIs, or honor proxy settings, this materially increases exposure to credential leakage, request redirection, and network boundary bypass.

Content

No source excerpt is available for this finding.

Known Vulnerable Dependency: basic-ftp==5.2.0 — 4 advisory(ies): GHSA-6v7q-wjvx-w8wg (basic-ftp: Incomplete CRLF Injection Protection Allows Arbitrary FTP Command Exe); CVE-2026-39983 (basic-ftp has FTP Command Injection via CRLF); CVE-2026-41324 (basic-ftp vulnerable to denial of service via unbounded memory consumption in Cl) +1 more

High
Category
Supply Chain
Confidence
95% confidence
Finding

basic-ftp 5.2.0 is affected by command injection and denial-of-service advisories, making it dangerous wherever attacker-controlled FTP paths, filenames, or server responses may be processed. Because this package appears in a broad agent/tooling dependency graph, any automated file retrieval or remote integration path would amplify the risk.

Content

No source excerpt is available for this finding.

Known Vulnerable Dependency: brace-expansion==5.0.5 — 4 advisory(ies): CVE-2026-13149 (brace-expansion: DoS via exponential-time expansion of consecutive non-expanding); CVE-2026-45149 (brace-expansion: Large numeric range defeats documented `max` DoS protection); CVE-2026-14257 (brace-expansion: DoS via unbounded expansion length causing an out-of-memory pro) +1 more

High
Category
Supply Chain
Confidence
95% confidence
Finding

brace-expansion 5.0.5 has multiple denial-of-service advisories involving exponential or unbounded expansion behavior. In tooling that processes glob patterns, ignore lists, archives, or user-supplied path expressions, this can be abused to consume CPU or memory and destabilize agent execution.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The README prominently advertises real-time archiving of complete conversation content, metadata, and operation records into a local SQLite database, but the privacy and retention risks are only mentioned later as a brief note. For a plugin that stores potentially sensitive prompts, tool outputs, credentials, file paths, and operational history, weak up-front disclosure can lead to accidental collection and long-term storage of confidential data without informed consent or retention controls.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The skill explicitly states that it automatically saves conversation messages to a SQLite database, but it does not warn users about the persistence of potentially sensitive content. This creates a privacy and data-handling risk because users may disclose secrets, personal data, or regulated information without realizing it is being retained locally.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

This code persistently stores full session messages, token usage, and operation history to a local SQLite database under the user's home directory, but there is no evidence in this component of consent, disclosure, retention controls, encryption, or access restriction. Because the stored data may include prompts, tool outputs, file paths, account identifiers, and operational history, compromise of the local host or unintended multi-user access could expose sensitive conversation and activity data.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
83% confidence
Finding

Operation audit records store detailed activity data such as session keys, operation type, target, serialized details, operator, and timestamps, but this file shows no access-control, minimization, or disclosure safeguards around that archival behavior. If operational details contain secrets, personal data, or sensitive workflow context, the archive becomes a secondary sensitive-data store that increases breach and insider-risk exposure.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
87% confidence
Finding

The engine persists full session content and rich metadata including content, account identifiers, message identifiers, tool names, media paths, and token usage without any visible consent, minimization, or redaction controls in this code path. In an agent/session archive context, this can expose sensitive user prompts, internal tool interactions, and attachment references if the database is accessed improperly or retained longer than intended.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The plugin explicitly states it will archive every conversation message to a local SQLite database, and the registration code enables this behavior without any visible consent, disclosure, scope restriction, or opt-in mechanism in this file. Because conversation data may include secrets, credentials, personal data, or proprietary content, silent comprehensive logging creates a meaningful privacy and data-retention risk.

Content

No source excerpt is available for this finding.

Ssd 3

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The plugin is designed for broad conversation logging and additionally attempts to register itself under the "default" context engine, with comments noting that multiple plugins using the same name will all trigger. In skill/plugin context, this makes the behavior more dangerous because it can expand collection scope beyond explicit user intent and cause pervasive capture of all conversations, including sensitive material, with little visibility to users.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
87% confidence
Finding

This code silently creates and uses a SQLite database under the user's home directory to persist session messages, operations, and token usage, including potentially sensitive content such as prompts, tool activity, account identifiers, and media paths. In an agent skill context, that increases privacy and confidentiality risk because users may not expect durable local storage of conversation and operational history, and the file may be exposed to other local processes, backups, or shared-account users if not additionally protected.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The engine persists full session messages plus metadata such as model, token counts, channel, accountId, messageId, toolName, mediaPath, and token usage without any indication here of consent, minimization, or redaction. In an agent context, messages may contain secrets, personal data, internal prompts, or tool outputs, so silent archival increases privacy and data-retention risk if the database is accessed, over-retained, or queried beyond user expectations.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

By default, the plugin stores archives in a persistent database under the user's home directory, creating long-lived local storage of potentially sensitive conversation data. In this skill context, local persistence without explicit warning or consent increases the chance that secrets remain on disk longer than expected and are later exposed through host compromise, backups, or shared accounts.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The plugin description states that every conversation message is archived to SQLite in real time, but this file shows no user-facing warning, consent flow, or disclosure mechanism. In a conversation-processing skill, silent comprehensive archiving is especially dangerous because it can capture sensitive prompts, secrets, personal data, and agent outputs without meaningful user awareness.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The plugin is marketed as a session archiver, but it also attempts to register itself as the global "default" context engine, which broadens its execution scope to conversations that did not explicitly opt into this engine. In this context, that means conversation data may be intercepted and persisted more widely than users or operators expect, increasing privacy and integrity risk.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

The skill content is presented entirely in Chinese, including headings and usage instructions, with no indication that the locale is optional or that the skill is intended only for a Chinese-speaking environment. This can violate language/locale policy where user-facing skills should not force a specific language without opt-in or justification.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
78% confidence
Finding

The inline comments use Chinese phrasing such as 公开方法:记录操作, and similar forced single-locale wording appears elsewhere in the file. Under the language/locale policy, natural-language content should not impose a specific language without opt-in or clear justification.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
80% confidence
Finding

The comment states that the session-archive engine is a 'pure side-effect engine' that never compacts context. However, this module also implements assemble() to return messages plus an estimated token count, which is active context-processing behavior rather than only side effects.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
81% confidence
Finding

The file includes natural-language comments in Chinese while the surrounding code and identifiers are otherwise English. This may conflict with organizational language policy if contributors or operators are expected to receive instructions in a default language without an explicit locale choice or documented justification.

Content

No source excerpt is available for this finding.

Known Vulnerable Dependency: @anthropic-ai/sdk==0.80.0 — 2 advisory(ies): CVE-2026-34451 (Claude SDK for TypeScript: Memory Tool Path Validation Allows Sandbox Escape to ); CVE-2026-41686 (Claude SDK for TypeScript has Insecure Default File Permissions in Local Filesys)

Low
Category
Supply Chain
Confidence
94% confidence
Finding

The lockfile includes @anthropic-ai/sdk version 0.80.0, which is explicitly flagged by advisories for sandbox escape via path validation and insecure default file permissions. Even though this file is only a dependency manifest, shipping a vulnerable pinned version is a real supply-chain risk if the package is used by the skill or its host environment.

Content

No source excerpt is available for this finding.

Known Vulnerable Dependency: @hono/node-server==1.19.11 — 2 advisory(ies): CVE-2026-39406 (@hono/node-server: Middleware bypass via repeated slashes in serveStatic); GHSA-frvp-7c67-39w9 (Node.js Adapter for Hono: Path traversal in `serve-static` on Windows via encode)

Low
Category
Supply Chain
Confidence
90% confidence
Finding

@hono/node-server 1.19.11 is listed with advisories for middleware bypass and Windows path traversal in static file serving. In a lockfile this is a real vulnerable dependency, though the practical risk depends on whether the affected serveStatic functionality is exposed by the skill or by transitive tooling.

Content

No source excerpt is available for this finding.

Known Vulnerable Dependency: @protobufjs/utf8==1.1.0 — 1 advisory(ies): CVE-2026-44288 (protobufjs has overlong UTF-8 decoding)

Low
Category
Supply Chain
Confidence
82% confidence
Finding

@protobufjs/utf8 1.1.0 is flagged for overlong UTF-8 decoding, which can enable input validation inconsistencies in systems relying on strict UTF-8 handling. This is a real dependency issue, but in the lockfile alone the impact is likely limited unless untrusted protobuf or text parsing is security-sensitive in the surrounding application.

Content

No source excerpt is available for this finding.

Known Vulnerable Dependency: body-parser==2.2.2 — 1 advisory(ies): CVE-2026-12590 (body-parser vulnerable to denial of service when invalid limit value silently di)

Low
Category
Supply Chain
Confidence
84% confidence
Finding

body-parser 2.2.2 is flagged for a denial-of-service condition tied to invalid limit handling. This is a real vulnerable dependency, though its severity remains low unless the skill exposes HTTP endpoints that parse attacker-controlled request bodies using affected defaults.

Content

No source excerpt is available for this finding.

Unpinned Dependencies

Low
Category
Supply Chain
Confidence
40% confidence
Finding

Dependencies lack version pinning, allowing potential malicious package updates. Consider pinning versions.

Content

Scanner excerpt · package.json (reported line 15)May include surrounding context.

json
"prepublishOnly": "npm run build"
  },
  "dependencies": {
    "better-sqlite3": "^11.0.0"
  },
  "devDependencies": {
    "@types/better-sqlite3": "^7.6.11",