Back to skill

Security audit

win-loss-review

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this skill suspicious with high confidence because scanners report an automatically downloaded, unpinned CLI and contact data collected beyond the stated need; one executable file was only partially inspected.

Consider avoiding installation until the CLI fallback is pinned to a reviewed version and contact data collection is limited to necessary fields. The scan covered 14 of 15 components fully; one executable file was partially inspected, so the evidence is incomplete.

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
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Memory PoisoningPersistent Context Injection, Context Window Stuffing, Memory Manipulation
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Findings (7)

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 129)May include surrounding context.

md
5. **Plan, then stop.** `node --import tsx evals/contract.mjs` from this skill's folder, then

Skill declares no tool scope ('permissions' or 'allowed-tools') but code capabilities were detected: shell.

Medium
Category
MCP Least Privilege
Confidence
70% confidence
Finding

Without declared permissions the skill's intent is opaque and cannot be validated.

Content

No source excerpt is available for this finding.

Context Window Stuffing

Medium
Category
Memory Poisoning
Confidence
80% confidence
Finding

Skill attempts to fill the context window with filler content, displacing legitimate instructions and safety constraints. This can degrade agent performance or bypass safety boundaries.

Content

Scanner excerpt · SKILL.md (reported line 152)May include surrounding context.

md
_asked_ genuinely live in the operator's head; the agent runs on a cron, so it asks in the pull
request body, never in a chat.

| Input                                                     | Kind      | How it is answered                                                                                                                                                                                                                                                                          | Why it matters                                                                                                                                                                                         |
| --------------------------------------------------------- | --------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Whitespace Padding

Medium
Category
Prompt Injection
Confidence
70% confidence
Finding

Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.

Content

Scanner excerpt · SKILL.md (reported line 160)May include surrounding context.

md
|
| Slack channel (`infra/agents/win-loss-analyst.ts`)        | value     | **asked**: the channel the digest lands in, locked on the `postMessage` use. An id (`C...`), not a name                                                                                                                                                                                       | Locked so the agent cannot pick a customer shared channel; a digest about lost deals is internal                                                                                                       |
| project binding (`infra/agents/win-loss-analyst.ts`)      | value     | **derived**: leave `state` unset and deploy binds the project's connected state (`cargo.state.json`): its repository, directory and production branch. `cargo-ai project check` prints the binding; a project whose state is not connected deploys the agent unbound, with an empty sandbox | This is the working tree the harnes

Whitespace Padding

Medium
Category
Prompt Injection
Confidence
70% confidence
Finding

Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.

Content

Scanner excerpt · SKILL.md (reported line 160)May include surrounding context.

md
| The disqualifier is the half of an ICP that protects the team's time, and won versus lost is the only place it comes from with evidence                                                                |
| Slack channel (`infra/agents/win-loss-analyst.ts`)        | value     | **asked**: the channel the digest lands in, locked on the `postMessage` use. An id (`C...`), not a name                                                                                                                                                                                       | Locked so the agent cannot pick a customer shared channel; a digest about lost deals is internal                                                                                                       |
| project binding (`infra/agents/win-loss-analyst.ts`)      | value     | **derived**: leave `state` unset and dep

MCP server referenced without pinned version: 'npx --yes @cargo-ai/cli'.

Medium
Category
MCP Rug Pull
Confidence
97% confidence
Finding

The fallback command automatically downloads and executes an unpinned npm package when the CLI is absent; it is a CLI execution risk, not specifically an MCP-server issue. A compromised or malicious future release could execute with the agent process's available repository access and credentials, and --yes removes the installation confirmation.

Content

No source excerpt is available for this finding.

The contact model extracts all fields despite documentation stating that only titles and IDs are read, never email.

Medium
Category
Excessive Agency
Confidence
93% confidence
Finding

The model persistently extracts every contact field on an hourly schedule, although this workflow needs only titles and association identifiers. Downstream queries avoiding email do not prevent emails and potentially sensitive custom properties from being copied into storage, increasing exposure if storage access is misconfigured or compromised; this is overcollection rather than evidence of external exfiltration.

Content

No source excerpt is available for this finding.