The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process content from Jira issues, which constitutes an untrusted ingestion point.
Ingestion points: The skill retrieves issue summaries, descriptions, and comments through the jira issue view CLI command and the mcp__atlassian__getJiraIssue MCP tool (referenced in SKILL.md and references/mcp.md).
Boundary markers: The instructions do not define specific delimiters or boundary markers to separate the fetched Jira content from the agent's system instructions, which could allow malicious instructions in a ticket to influence the agent's behavior.
Capability inventory: The skill possesses significant capabilities, including executing shell commands via the jira CLI and a local jira-link-issues script, and performing write operations like creating, updating, and transitioning tickets.
Sanitization: No sanitization or filtering logic is specified for data retrieved from Jira before it is processed or used in subsequent tool calls.
[COMMAND_EXECUTION]: The skill relies on shell command execution for its primary functionality.
As documented in SKILL.md and references/commands.md, the skill executes the jira CLI for viewing, listing, and modifying issues.
It uses shell patterns such as writing content to temporary files (/tmp/jira_body.md) and command substitution (e.g., $(jira me)) to handle multi-line input and dynamic arguments.
A local shell script ~/.claude/skills/jira/jira-link-issues is invoked to handle issue linking, which is outside the standard MCP toolset.

jira