The Agent Skills Directory

[DATA_EXFILTRATION]: The skill performs network operations to the official Slack API (slack.com) to manage messages, channels, and files. These operations are essential for the skill's stated purpose and utilize the user-provided $SLACK_TOKEN environment variable for authentication.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external sources, specifically Slack message history and user search results, which could potentially contain adversarial content.
Ingestion points: Data is retrieved via conversations.list, search.messages, and users.lookupByEmail API calls in SKILL.md.
Boundary markers: No specific delimiters or "ignore instructions" markers are used for processed message content.
Capability inventory: The skill utilizes shell execution via curl and jq as documented in SKILL.md.
Sanitization: The skill correctly employs jq --arg for building JSON payloads and curl --data-urlencode for query parameters, which effectively mitigates the risk of shell command injection from untrusted input.

slack