The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through the ingestion of external data.
Ingestion points: The read-issue.py script fetches comments and attachments from Linear issues, which are then presented to the agent.
Boundary markers: There are no explicit delimiters or instructions to ignore embedded commands within the fetched content.
Capability inventory: The skill possesses several sensitive capabilities, including file uploads via upload-file.py, issue state transitions via move-issue.py, and linking external URLs/PRs via attach-pr.py and attach-url.py.
Sanitization: No sanitization or filtering is performed on the content retrieved from Linear, allowing potentially malicious instructions in comments to influence agent behavior.
[DATA_EXFILTRATION]: The upload-file.py utility can be used to upload any local file the agent has access to. The script explicitly sets the makePublic flag to True when requesting an upload URL from Linear, meaning any file uploaded will be accessible via a public URL provided by the service. This could lead to accidental or malicious exposure of sensitive data if the agent is manipulated into uploading protected files.

linear