The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes untrusted data from Jira issue fields, which could contain malicious instructions designed to influence agent behavior (Indirect Prompt Injection).
Ingestion points: jira-issue.py and jira-search.py retrieve and display content (summaries, descriptions, comments) from external Jira instances.
Boundary markers: The skill does not implement specific delimiters or safety instructions to help the agent distinguish between ingested content and system instructions.
Capability inventory: The skill possesses extensive write capabilities, including the ability to create and update issues, transition statuses, and add comments via scripts such as jira-create.py and jira-update.py.
Sanitization: No sanitization or content filtering is performed on the data retrieved from Jira before it is presented to the agent.
[EXTERNAL_DOWNLOADS]: The skill includes functionality for downloading external files and recommends tool installation from remote sources.
jira-attachment.py allows downloading Jira attachments to the local filesystem using the requests library.
Documentation in README.md and error messages in jira-validate.py recommend installing the uv tool via a piped shell script from astral.sh, which is a well-known and reputable service in the developer community.
[COMMAND_EXECUTION]: The skill executes local system commands as part of its environment validation process.
jira-validate.py runs uv --version using subprocess.run to verify the presence and version of the required package manager.

jira-integration