Security audit

SummitScribe

Security checks for vulnerabilities and agentic risk

Overview

ClawScan classifies the SummitScribe skill as benign with high confidence: the completed scans found no AIG issues, and the two Skillspector alerts were judged expected in context of its setup and safety guidance.

The verdict is benign, but that does not mean risk-free. Before installing, consider whether the listed SummitScribe read scopes suit your needs; keep any access token in a secret store and do not paste it into chat. Skillspector reports complete coverage of the single Markdown component, and AIG completed with no findings.

A.I.G

Vulnerability Patterns

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
Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns

Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (2)

Credential Access

High

Category: Privilege Escalation
Confidence: 70% confidence
Finding: Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.
Content: Scanner excerpt · SKILL.md (reported line 31)May include surrounding context.
md
## Setup and safety - The server is `https://summitscribe.app/api/mcp`. It needs either a browser sign-in (OAuth, for tools that support it) or a SummitScribe access token. Either way the connection needs the `mcp:read` scope plus a data scope for each tool (`meetings:read`, `notes:read`, `transcripts:read`, `actions:read`, `search:read`). - The user creates tokens, and disconnects signed-in apps, at summitscribe.app in Account settings, then Developer & agents. Keep the token in `SUMMITSCRIBE_TOKEN` or your tool's secret store. Never ask the user to paste it into the chat, and never print it. - A 401 means the sign-in or token expired or was revoked; ask the user to sign in again or make a new token on the web. A 403 names the missing scope. A 429 means wait and retry later. - If the user thinks a token leaked, tell them to revoke it at summitscribe.app right away.

Autonomous Decision Making

Medium

Category: Excessive Agency
Confidence: 80% confidence
Finding: Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.
Content: Scanner excerpt · SKILL.md (reported line 32)May include surrounding context.
md
## Setup and safety - The server is `https://summitscribe.app/api/mcp`. It needs either a browser sign-in (OAuth, for tools that support it) or a SummitScribe access token. Either way the connection needs the `mcp:read` scope plus a data scope for each tool (`meetings:read`, `notes:read`, `transcripts:read`, `actions:read`, `search:read`). - The user creates tokens, and disconnects signed-in apps, at summitscribe.app in Account settings, then Developer & agents. Keep the token in `SUMMITSCRIBE_TOKEN` or your tool's secret store. Never ask the user to paste it into the chat, and never print it. - A 401 means the sign-in or token expired or was revoked; ask the user to sign in again or make a new token on the web. A 403 names the missing scope. A 429 means wait and retry later. - If the user thinks a token leaked, tell them to revoke it at summitscribe.app right away.