Back to skill

Security audit

Opentwitter

Security checks for vulnerabilities and agentic risk

Overview

The scanners report a benign skill for retrieving Twitter/X and meme data, with medium confidence; the main caveat is that it also documents changes to persistent monitoring lists, beyond the stated retrieval purpose.

The verdict is benign with medium confidence, not a guarantee of safety. Before installing, consider whether you want a data-retrieval skill that also documents adding and deleting monitoring entries. Coverage is incomplete: Skillspector scanned both listed components but marked analysis partial because some local path-like references could not be resolved; the AIG scan completed with no findings.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (14)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

The twitter_watch_add endpoint and its configuration fields indicate a state-changing operation that adds an account to monitoring. Although the service and subject matter match the description, managing watch entries is a distinct capability not disclosed by the declared data-access purpose.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% 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 · SKILL.md (reported line 12)May include surrounding context.

md
env:
        - OPENNEWS_TOKEN
      bins:
        - curl
    primaryEnv: OPENNEWS_TOKEN
    emoji: "\U0001F426"
    install:

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% 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 · SKILL.md (reported line 123)May include surrounding context.

Search with query:

bash
curl -s -X POST "https://ai.6551.io/open/twitter_search" \
  -H "Authorization: Bearer $OPENNEWS_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"query": "from:VitalikButerin bitcoin -filter:replies", "maxResults": 20, "product": "Top"}'

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% 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 · SKILL.md (reported line 181)May include surrounding context.

bash
# Get new followers
curl -s -X POST "https://ai.6551.io/open/twitter_follower_events" \
  -H "Authorization: Bearer $OPENNEWS_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"username": "elonmusk", "isFollow": true, "maxResults": 20}'

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% 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 · SKILL.md (reported line 204)May include surrounding context.

Get deleted tweets from a user.

bash
curl -s -X POST "https://ai.6551.io/open/twitter_deleted_tweets" \
  -H "Authorization: Bearer $OPENNEWS_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"username": "elonmusk", "maxResults": 20}'

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% 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 · SKILL.md (reported line 220)May include surrounding context.

Get which KOLs (Key Opinion Leaders) are following a user.

bash
curl -s -X POST "https://ai.6551.io/open/twitter_kol_followers" \
  -H "Authorization: Bearer $OPENNEWS_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"username": "elonmusk"}'

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% 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 · SKILL.md (reported line 235)May include surrounding context.

Get Twitter article by ID.

bash
curl -s -X POST "https://ai.6551.io/open/twitter_article_by_id" \
  -H "Authorization: Bearer $OPENNEWS_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"id": "article_id"}'

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% 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 · SKILL.md (reported line 252)May include surrounding context.

This endpoint retrieves a tweet by ID and automatically fetches any tweets it replies to or quotes, providing complete context.

bash
curl -s -X POST "https://ai.6551.io/open/twitter_tweet_by_id" \
  -H "Authorization: Bearer $OPENNEWS_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"twId": "2030318958512164966"}'

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% 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 · SKILL.md (reported line 272)May include surrounding context.

Get tweets that quote a specific tweet.

bash
curl -s -X POST "https://ai.6551.io/open/twitter_quote_tweets_by_id" \
  -H "Authorization: Bearer $OPENNEWS_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"id": "2030318958512164966", "maxResults": 20}'

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% 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 · SKILL.md (reported line 288)May include surrounding context.

Get users who retweeted a specific tweet.

bash
curl -s -X POST "https://ai.6551.io/open/twitter_retweet_users_by_id" \
  -H "Authorization: Bearer $OPENNEWS_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"id": "2030318958512164966"}'

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% 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 · SKILL.md (reported line 304)May include surrounding context.

Get all Twitter monitoring users for the current user.

bash
curl -s -X POST "https://ai.6551.io/open/twitter_watch" \
  -H "Authorization: Bearer $OPENNEWS_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{}'

The skill exposes persistent watch-list mutations despite describing only Twitter/X and meme data retrieval.

Medium
Category
Excessive Agency
Confidence
97% confidence
Finding

The manifest at L003 describes data-access capabilities, but L315-L318 provides an authenticated request to add monitoring entries, and L344-L347 provides one to delete them. Adding an entry can also create a watch group (L324) and configure ongoing monitoring, so these operations change remote account state rather than merely retrieve the advertised data.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% 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 · SKILL.md (reported line 344)May include surrounding context.

Delete a Twitter user from monitoring list by username.

bash
curl -s -X POST "https://ai.6551.io/open/twitter_watch_delete" \
  -H "Authorization: Bearer $OPENNEWS_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"username": "elonmusk"}'

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% 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 · SKILL.md (reported line 556)May include surrounding context.

Crypto Twitter KOL Tweets

bash
curl -s -X POST "https://ai.6551.io/open/twitter_user_tweets" \
  -H "Authorization: Bearer $OPENNEWS_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"username": "VitalikButerin", "maxResults": 10}'