linkedin-comment-writer
Pass
Audited by Gen Agent Trust Hub on Oct 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input from LinkedIn posts and comments, creating a potential surface for indirect prompt injection attacks.
- Ingestion points: The skill accepts external LinkedIn post text and existing comments as data inputs, as defined in
SKILL.mdand the brief template inassets/comment_brief_template.md. - Boundary markers:
SKILL.mdprovides explicit instructions for the agent to treat pasted text as data and ignore any embedded instructions. It also defines a 'NOTICE' reporting mechanism for when such attempts are detected. - Capability inventory: The skill has the capability to generate text and execute a local validation script (
scripts/comment_linter.py) via the CLI. The Python scripts themselves do not initiate further subprocesses, network operations, or file system writes. - Sanitization: The
scripts/comment_rules.pyfile implementsINJECTION_PATTERNS, a set of regular expressions designed to identify and flag prompt injection techniques (e.g., 'ignore previous instructions') in the input data.
Audit Metadata