The Agent Skills Directory

[PROMPT_INJECTION]: Metadata Inconsistency: The SKILL.json file identifies the author as 'Anthropic Marketplace 3p', which contradicts the provided author context '420company'. This misleading branding could lead users to trust the skill more than appropriate.
[PROMPT_INJECTION]: Indirect Prompt Injection: The skill processes untrusted data from Jira issue summaries and descriptions (Ingestion points: SKILL.md, Step 2/3). It lacks explicit boundary markers to delimit this untrusted content. The skill has capabilities to create or update Confluence pages and query Jira (Capability inventory: SKILL.md, Step 2/5). While scripts/jql_builder.py provides sanitization for JQL queries, it does not sanitize the content formatted into the Confluence report.
[SAFE]: The skill uses standard Atlassian tools for their intended project management purposes.
[SAFE]: The inclusion of scripts/jql_builder.py is a security best practice, providing a regex-based allow-list to prevent JQL injection attacks.

generate-status-report