implement-issue-tree

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). The required workflow for implement-issue-tree monitors and consumes outsider-authored free text from GitHub issues, pull requests, and review comments via the gh CLI and REST/GraphQL APIs.

MEDIUM W009: Direct money access capability detected (payment gateways, crypto, banking).

• Direct money access detected (medium risk: 0.30). The skill documents operational workflows that instruct an agent to execute GitHub Actions, issue creation, branch pushes, pull request creation, and pull request merging (including squash merging via gh pr merge) against real repositories. This constitutes a direct financial-security or operational-asset exposure (managing operational infrastructure and code release pipelines). However, the skill explicitly enforces that autonomous client-side merges only occur under strict conditions (autoMerge: true combined with explicitly defined externalChecks), and otherwise stops at a blocked/human-review state. Crucially, the execution layer relies on independent GitHub branch protection rulesets and repository permissions (server-enforcement, rulesets, and required status checks) to authorize or reject these actions independently of the agent's instructions, qualifying it for a medium tier under direct financial/operational actions with independent service-layer controls.