ClawHub
Back to skill

Security audit

Axion

Security checks for vulnerabilities and agentic risk

Overview

This is a legitimate Axion forecasting helper, but its bundled API reference exposes account actions like public sharing, deletion, and credit purchase without explicit confirmation guidance.

Install only if you are comfortable sending forecast questions and related context to Axion using your API key. Treat any request to share, delete, stop, list account history, or purchase credits as requiring a separate explicit confirmation, since the skill itself does not define those guardrails.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (6)

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
94% confidence
Finding

The API reference exposes high-risk actions including thread deletion, public sharing, cancellation with irreversible credit consumption, and credit purchase, but provides no safety guidance or constraints around who may trigger them. In an agent skill, these parameters and endpoints are especially dangerous because a prompt-driven workflow could be manipulated into destructive, privacy-impacting, or billable actions without clear authorization.

Content

Scanner excerpt · references/axion-api.md (reported line 57)May include surrounding context.

md
- `GET /forecasts`: list all threads for the account.
- `POST /forecasts/{thread_id}/stop`: cancel an in-progress forecast (consumed credits still charged).
- `POST /forecasts/{thread_id}/share` / `.../unshare`: toggle public visibility; share returns `{ "share_url": "/share/..." }`.
- `DELETE /forecasts/{thread_id}`: delete a thread.
- `GET /account/balance`: `{ "credits": 3750 }`.
- `POST /account/credits/purchase`: `{ "amount": 50 }` returns `{ "checkout_url": "https://checkout.stripe.com/..." }`. Minimum $50.

External Transmission

Medium
Category
Data Exfiltration
Confidence
89% confidence
Finding

The documented endpoint identifies a third-party external destination for outbound network traffic. While not inherently malicious, hardcoded external API use expands the trust boundary and can expose user prompts, business context, or metadata to an outside provider as part of normal operation.

Content

Scanner excerpt · SKILL.md (reported line 37)May include surrounding context.

  1. Start a forecast:
bash
curl -s https://api.axion.eternis.ai/forecasts \
  -H "Authorization: Bearer $AXION_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"input": "Will the Fed cut rates in June 2026?", "effort": "high"}'

External Transmission

Medium
Category
Data Exfiltration
Confidence
89% confidence
Finding

The documented endpoint identifies a third-party external destination for outbound network traffic. While not inherently malicious, hardcoded external API use expands the trust boundary and can expose user prompts, business context, or metadata to an outside provider as part of normal operation.

Content

Scanner excerpt · SKILL.md (reported line 37)May include surrounding context.

  1. Start a forecast:
bash
curl -s https://api.axion.eternis.ai/forecasts \
  -H "Authorization: Bearer $AXION_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"input": "Will the Fed cut rates in June 2026?", "effort": "high"}'

External Transmission

Medium
Category
Data Exfiltration
Confidence
86% confidence
Finding

The polling request continues transmitting the bearer token to the same external service and may leak workflow metadata such as forecast IDs and timing. This is lower impact than the initial submission because it typically sends less user content, but it still represents ongoing outbound communication to a third party.

Content

Scanner excerpt · SKILL.md (reported line 48)May include surrounding context.

  1. Poll until status is completed or failed, backing off a few seconds between calls:
bash
curl -s https://api.axion.eternis.ai/forecasts/<id> \
  -H "Authorization: Bearer $AXION_API_KEY"

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The reference documents destructive and privacy-affecting endpoints such as share/unshare, stop, delete, and credit purchase without any explicit warnings about user consent, billing effects, or public exposure. In an agent skill context, this increases the chance that an agent will invoke these actions automatically or without sufficient confirmation, leading to unwanted deletion, charges, or data disclosure.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
70% 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 · references/axion-api.md (reported line 88)May include surrounding context.

md
BASE = "https://api.axion.eternis.ai"
headers = {"Authorization": f"Bearer {API_KEY}"}

r = requests.post(f"{BASE}/forecasts", headers=headers,
                  json={"input": "Will the Fed cut rates in June 2026?", "effort": "high"})
thread_id = r.json()["id"]

Static analysis

No suspicious patterns detected.