Back to skill

Security audit

pocket-recordings-transcripts

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed AgentPMT connector for Pocket recordings, but it can access, save, upload, and link sensitive audio and transcript content.

Install only if you intend an agent to work with your Pocket recordings through AgentPMT. Treat transcripts and audio as sensitive: confirm authorization before retrieving recordings, generating download links, saving audio copies, or uploading File Manager audio for transcription, and avoid running unpinned npx setup commands in environments with unnecessary secrets.

Vulnerability Patterns
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • 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)

T08 · Insecure Dependencies

Warning
Location
SKILL.md:170
Finding

Unpinned Package and Repository Execution During Skill Setup

Content
View full analysis

Vulnerability Details

File Location: SKILL.md, lines 170-180
Vulnerability Type: Unpinned third-party package and mutable repository dependency
Risk Level: Medium

Vulnerable Code

markdown
- AgentPMT account MCP/REST setup: ../agentpmt-account-mcp-rest-api-setup
  - ClawHub page: https://clawhub.ai/agentpmt/agentpmt-account-mcp-rest-api-setup
  - OpenClaw install: `openclaw skills install agentpmt-account-mcp-rest-api-setup`
  - skills.sh install: `npx skills add AgentPMT/agent-skills --skill agentpmt-account-mcp-rest-api-setup`

skills.sh install script:

```bash
npx skills add AgentPMT/agent-skills --skill what-is-agentpmt
npx skills add AgentPMT/agent-skills --skill agentpmt-account-mcp-rest-api-setup
text

Equivalent unpinned commands also appear at `SKILL.md:82` and `SKILL.md:229-230`.

### Technical Analysis

The documented setup process invokes the unversioned `skills` npm package through `npx` and installs content from `AgentPMT/agent-skills` without specifying an immutable commit, release, checksum, or signature. Consequently, the code and Skill instructions obtained during a future installation may differ from the content that was originally reviewed.

Because `npx` can download and execute package content, compromise of the npm package, its dependency graph, the referenced repository, or the associated publishing accounts could turn this setup step into a code-execution or instruction-injection channel. This is a supply-chain weakness rather than evidence that the currently reviewed files contain an active malicious payload.

### Attack Path

1. An attacker compromises the unversioned `skills` npm package, one of its transitive dependencies, the referenced `AgentPMT/agent-skills` repository, or a relevant publisher account.
2. The attacker publishes malicious package code or replaces the remotely retrieved Skill content.
3. A user or Agent follows the documented `npx skills add ...` setup command.
4. `npx` retrieves and ex
...[truncated 1029 chars]
Remediation
View remediation

Remediation Suggestions

  1. Pin the skills npm package to a specifically reviewed version rather than relying on the latest resolution:
    bash
    npx --yes skills@<reviewed-version> add AgentPMT/agent-skills --skill what-is-agentpmt
    
  2. Pin AgentPMT/agent-skills to an immutable commit hash or cryptographically signed release if the installer supports repository revisions.
  3. Publish and verify SHA-256 checksums or signatures for downloaded Skill artifacts before installation.
  4. Use a lockfile and integrity-protected package installation process for the installer and all transitive dependencies.
  5. Prefer a reviewed local installation artifact over executing mutable content directly from the network.
  6. Run installation in a sandbox or least-privilege environment without unnecessary secrets, production credentials, or writable sensitive directories.
  7. Document the expected package version, repository commit, checksums, publisher identity, and verification procedure in SKILL.md.
  8. Re-audit pinned installer code and installed Skill content whenever the pinned versions are updated.
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (12)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The skill handles highly sensitive content—recordings, transcripts, AI summaries, downloadable audio, and uploads—but does not prominently require user confirmation, consent checks, or privacy warnings before accessing or moving that data. In a voice/meeting context, this increases the risk of exposing confidential conversations, regulated data, or third-party information through routine agent use.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 232)May include surrounding context.

md
- What AgentPMT is: ../what-is-agentpmt (ClawHub: `what-is-agentpmt`, page: https://clawhub.ai/agentpmt/what-is-agentpmt; skills.sh: `npx skills add AgentPMT/agent-skills --skill what-is-agentpmt`)
- AgentPMT account MCP/REST setup: ../agentpmt-account-mcp-rest-api-setup (ClawHub: `agentpmt-account-mcp-rest-api-setup`, page: https://clawhub.ai/agentpmt/agentpmt-account-mcp-rest-api-setup; skills.sh: `npx skills add AgentPMT/agent-skills --skill agentpmt-account-mcp-rest-api-setup`)
- Marketplace product: https://www.agentpmt.com/marketplace/pocket-recordings-transcripts
- AgentPMT main MCP server: https://api.agentpmt.com/mcp/
- AgentPMT REST invoke endpoint: https://api.agentpmt.com/products/purchase

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 233)May include surrounding context.

md
- What AgentPMT is: ../what-is-agentpmt (ClawHub: `what-is-agentpmt`, page: https://clawhub.ai/agentpmt/what-is-agentpmt; skills.sh: `npx skills add AgentPMT/agent-skills --skill what-is-agentpmt`)
- AgentPMT account MCP/REST setup: ../agentpmt-account-mcp-rest-api-setup (ClawHub: `agentpmt-account-mcp-rest-api-setup`, page: https://clawhub.ai/agentpmt/agentpmt-account-mcp-rest-api-setup; skills.sh: `npx skills add AgentPMT/agent-skills --skill agentpmt-account-mcp-rest-api-setup`)
- Marketplace product: https://www.agentpmt.com/marketplace/pocket-recordings-transcripts
- AgentPMT main MCP server: https://api.agentpmt.com/mcp/
- AgentPMT REST invoke endpoint: https://api.agentpmt.com/products/purchase

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The upload_recording action sends a File Manager audio file to Pocket for transcription, but the schema does not clearly warn that the audio is transmitted to an external service for processing. Because recordings may contain sensitive conversations or regulated data, lack of explicit disclosure can lead to unauthorized third-party sharing and privacy/compliance violations.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
91% confidence
Finding

The save_audio_to_files action persists downloaded recording audio into File Manager, but the schema text does not explicitly warn that this creates a stored copy of potentially sensitive meeting audio. In an agent setting, unclear disclosure can cause users or downstream agents to trigger durable storage of confidential recordings without informed consent, increasing privacy and data-retention risk.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.