fix: Skill 未引导 Agent 使用正确的部署工具路径，导致 HTTP 云函数创建后实际未生效

[binggg]

Attribution issue

• issueId: issue_mnzgm6vs_zs4kzk
• category: skill
• canonicalTitle: Skill 未引导 Agent 使用正确的部署工具路径，导致 HTTP 云函数创建后实际未生效
• representativeRun: atomic-js-cloudbase-http-function-basic/2026-04-15T02-49-54-801ec1

Automation summary

• root_cause: Real and fixable in this repo. The failing run used the all-in-one CloudBase guidance, not scattered MCP skills, and that guidance only treated exports.main(...) style Node.js functions as direct manageFunctions candidates. That misrouted HTTP cloud function work away from the correct deployment path. In addition, the cloud-functions skill did not make the post-create HTTP delivery steps explicit enough: HTTP functions still need explicit manageGateway(action="createAccess") for URL access, and legacy writeSecurityRule requests must be translated to queryPermissions / managePermissions rather than copied literally.
• changes: Updated config/source/skills/cloud-functions/SKILL.md to explicitly warn that createFunction alone does not make an HTTP function reachable, to require gateway + permission follow-up for externally accessible HTTP functions, and to map legacy names like writeSecurityRule to managePermissions. Updated config/source/guideline/cloudbase/SKILL.md and config/source/editor-config/guides/cloudbase-rules.mdc so the all-in-one deployment workflow now distinguishes Event vs HTTP functions, recognizes HTTP-function criteria (`port

Evaluation contract

• eval_scope: primary_only
• primary_cases:
  • atomic-js-cloudbase-http-function-basic
• regression_cases:
  • (none)

Changed files

• .agents/skills/mcp-attribution-worktree
• .agents/skills/mcp-attribution-worktree/SKILL.md
• .agents/skills/mcp-attribution-worktree/references/evaluation-verification.md
• .agents/skills/mcp-attribution-worktree/references/iteration-loop.md
• .agents/skills/mcp-attribution-worktree/references/report-api-workflow.md
• .agents/skills/mcp-attribution-worktree/references/subagent-orchestration.md
• .agents/skills/mcp-attribution-worktree/references/value-triage.md
• .agents/skills/mcp-attribution-worktree/references/worktree-repair.md
• .codebuddy/attribution-context/issue_mnzgm6vs_zs4kzk/README.md
• .codebuddy/attribution-context/issue_mnzgm6vs_zs4kzk/evaluation-trace.json
• .codebuddy/attribution-context/issue_mnzgm6vs_zs4kzk/issue.json
• .codebuddy/attribution-context/issue_mnzgm6vs_zs4kzk/representative-evidence.json
• .codebuddy/attribution-context/issue_mnzgm6vs_zs4kzk/run-result.json
• .codebuddy/attribution-context/issue_mnzgm6vs_zs4kzk/run-trace.json
• .codebuddy/skills/mcp-attribution-worktree
• .codebuddy/skills/mcp-attribution-worktree/SKILL.md
• .codebuddy/skills/mcp-attribution-worktree/references/evaluation-verification.md
• .codebuddy/skills/mcp-attribution-worktree/references/iteration-loop.md
• .codebuddy/skills/mcp-attribution-worktree/references/report-api-workflow.md
• .codebuddy/skills/mcp-attribution-worktree/references/subagent-orchestration.md
• .codebuddy/skills/mcp-attribution-worktree/references/value-triage.md
• .codebuddy/skills/mcp-attribution-worktree/references/worktree-repair.md
• config/source/editor-config/guides/cloudbase-rules.mdc
• config/source/guideline/cloudbase/SKILL.md
• config/source/skills/cloud-functions/SKILL.md
• tests/skill-quality-standards.test.js

[chatgpt-codex-connector]

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.