Back to skill

Security audit

agentfeed-briefs

Security checks for vulnerabilities and agentic risk

Overview

The skill is marked suspicious because scanners flagged an unpinned installation command and behavior outside its stated market-brief purpose; the scan was partial, so the evidence is incomplete.

Consider avoiding installation until the package and repository versions used by the setup command are pinned and reviewed. The scan covered only part of the artifact, so unresolved content could change the assessment.

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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (5)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

The supplied commands concern skill installation and distribution, not retrieving or summarizing AgentFeed market data. Publication is particularly outside the declared purpose. The comment indicates that publication has not occurred, so this assessment identifies an undeclared operation in the snippet rather than asserting successful execution.

Content

No source excerpt is available for this finding.

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 28)May include surrounding context.

md
eter; the signal routes take the optional ones listed in [references/ROUTES.md](references/ROUTES.md) and in `/openapi.json`.

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 112)May include surrounding context.

md
ne. Facts about the routes, prices and pass terms are in [references/ROUTES.md](references/ROUTES.md); they are generated from the AgentFeed code and change wit

Credential Access

High
Category
Privilege Escalation
Confidence
60% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · references/SCHEDULING.md (reported line 19)May include surrounding context.

Cron jobs in Hermes never prompt: a skill whose secret is missing is not asked for in chat. Put the key where Hermes reads secrets, then install the skill:

bash
echo 'AGENTFEED_KEY=afk_...' >> ~/.hermes/.env      # set by the user, once
hermes skills install seekdaseek/agentfeed-skills/skills/agentfeed-briefs

MCP server referenced without pinned version: 'npx skills'.

Medium
Category
MCP Rug Pull
Confidence
88% confidence
Finding

The suggested npx skills add command can download and execute an unpinned npm package, exposing anyone following it to compromised or unexpectedly changed upstream code with their local privileges. This is an installation supply-chain weakness, not an MCP-server configuration issue; its placement in optional setup documentation limits exposure but does not remove the risk.

Content

No source excerpt is available for this finding.