ClawHub
Back to skill

Security audit

Ccp Session Update

Security checks for vulnerabilities and agentic risk

Overview

The skill is mostly coherent as a session-continuity workflow, but it asks for broad persistence, password-manager credential access, and privileged Plane database mutation that users should review carefully.

Install only if you trust this workflow with your project summaries, Notion workspace, synced Obsidian vault, 1Password access, Docker daemon, and Plane database. Prefer replacing the 1Password command with an injected secret, adding redaction/confirmation before cloud sync, and using a supported Plane API or narrowly scoped helper instead of the Django shell.

A.I.G

Vulnerability Patterns
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
  • 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
Findings (2)

T05 · Unauthorized Access and Privilege Escalation

Error
Location
SKILL.md:164
Finding

Plane Authorization and Validation Bypass Through Privileged Django Shell Access

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
SKILL.md:89
Finding

Undeclared Password-Manager Credential Retrieval and Secret Exposure

Content
View full analysis
Remediation
View remediation

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Memory PoisoningPersistent Context Injection, Context Window Stuffing, Memory Manipulation
  • Rogue AgentSelf-Modification, Session Persistence
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (5)

Memory Manipulation

High
Category
Memory Poisoning
Confidence
80% confidence
Finding

Skill manipulates agent memory, state, or stored context. Memory corruption can alter personality, override safety rules, or cause unpredictable behavior.

Content

Scanner excerpt · SKILL.md (reported line 61)May include surrounding context.

md
_Last updated: YYYY-MM-DD HH:MM AEST_

## RESUME FROM HERE
[Single clear statement of the true next action — not a list]

## Live URLs
[All deployed endpoints, GitHub, devnet addresses]

Session Persistence

Medium
Category
Rogue Agent
Confidence
83% confidence
Finding

The skill deliberately persists session context across multiple stores: local project files, cloud Notion, synced Obsidian vault, and a task tracker. While this supports continuity, it also increases the blast radius of any sensitive data captured during a session, making accidental long-term retention and cross-system propagation more likely.

Content

Scanner excerpt · SKILL.md (reported line 39)May include surrounding context.

md
| **Obsidian Vault** | Proton Drive (`~/Library/CloudStorage/ProtonDrive-monkfenix@proton.me-folder/Vault/`) | Personal knowledge base — daily note + per-project file |
| **Plane** | Local Docker (`localhost:8086`) | Task tracking — create issues for new work, close completed ones |

**Always write STATUS.md before spawning Liv.** It's the source Notion and Obsidian summarise from. Race condition if you write it after.

---

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill explicitly retrieves a live Notion API credential and immediately uses it for an outbound API request, but it provides no warning about secret exposure, operator confirmation, or limits on what session data may be transmitted. In an agent setting, this increases the risk of unintended disclosure of sensitive project context or accidental misuse of a production credential.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
89% confidence
Finding

The presence of a hardcoded external destination indicates the skill is designed to transmit information outside the local environment. In the context of automated agent execution, this is dangerous because it normalizes network egress without verifying that the data being posted is safe or intended for third-party storage.

Content

Scanner excerpt · SKILL.md (reported line 98)May include surrounding context.

Create a log entry:

bash
curl -s -X POST https://api.notion.com/v1/pages \
  -H "Authorization: Bearer $NOTION_KEY" \
  -H "Notion-Version: 2022-06-28" \
  -H "Content-Type: application/json" \

External Transmission

Medium
Category
Data Exfiltration
Confidence
89% confidence
Finding

The presence of a hardcoded external destination indicates the skill is designed to transmit information outside the local environment. In the context of automated agent execution, this is dangerous because it normalizes network egress without verifying that the data being posted is safe or intended for third-party storage.

Content

Scanner excerpt · SKILL.md (reported line 98)May include surrounding context.

Create a log entry:

bash
curl -s -X POST https://api.notion.com/v1/pages \
  -H "Authorization: Bearer $NOTION_KEY" \
  -H "Notion-Version: 2022-06-28" \
  -H "Content-Type: application/json" \

Static analysis

No suspicious patterns detected.