champion-identifier-skill
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external LinkedIn profile data, creating a surface for indirect prompt injection.
- Ingestion points: LinkedIn profile text provided during the analysis workflow (SKILL.md).
- Boundary markers: Absent; there are no instructions to the agent to treat external content as untrusted or to use delimiters.
- Capability inventory: The skill contains instructions to write files to the local filesystem.
- Sanitization: No sanitization or filtering of the external profile data is performed.
- [COMMAND_EXECUTION]: The skill workflow includes an automated instruction to write a JSON file to a hidden directory in the user's home folder.
- Evidence: The 'Emit Outcome Sidecar' section in SKILL.md requires writing to
~/.claude/skill-analytics/last-outcome-champion-identifier.json.
Audit Metadata