skills/acedatacloud/skills/mastodon/Gen Agent Trust Hub

mastodon

Pass

Audited by Gen Agent Trust Hub on Aug 16, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: Employs standard CLI utilities including curl, jq, and uuidgen to perform REST API operations and process JSON responses. This usage is consistent with the skill's defined purpose.
  • [SAFE]: Demonstrates secure practices by utilizing platform-injected credentials ($MASTODON_ACCESS_TOKEN and $MASTODON_BASE_URL) instead of hardcoding sensitive information.
  • [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection. 1. Ingestion points: Data is retrieved from the Mastodon REST API via curl in SKILL.md (e.g., account verification and status listing). 2. Boundary markers: No explicit delimiters or instructions are provided to ignore embedded content in API responses. 3. Capability inventory: Includes network access (curl) and artifact publishing (publish_artifact). 4. Sanitization: Uses jq to parse structured data, which limits injection to the values of the JSON fields themselves.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 16, 2026, 12:55 PM
Security Audit — agent-trust-hub — mastodon