The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill utilizes npx @web42/stask which downloads and executes code from a third-party npm package at runtime. This package is not associated with a known trusted organization or the author's identified namespace (yarn-rp).
[COMMAND_EXECUTION]: The agent is instructed to run shell commands using npx and the GitHub CLI (gh) to perform task coordination, pull request management, and environment setup.
[PROMPT_INJECTION]: The workflow is susceptible to indirect prompt injection because it ingests untrusted data from external sources and uses it to drive agent behavior without adequate safeguards.
Ingestion points: The agent reads task specifications from Slack files (retrieved via show) and pull request feedback (retrieved via pr-status).
Boundary markers: No delimiters or instructions are provided to treat data from these external sources as untrusted or to ignore instructions embedded within them.
Capability inventory: The agent has permissions to create subtasks, transition task statuses, create and update GitHub pull requests, and post messages to Slack via chat.postMessage.
Sanitization: There is no evidence of sanitization, validation, or filtering of content retrieved from Slack or GitHub before it is processed by the agent.

stask-lead