Skill Details
Runtime
(claude-4-7-opus)
Example user message
[userMessage]: Add a new organization-level toggle that hides the analytics section in the admin panel and disables the related nightly summarization job. Touch backend, web, and the pipeline service.
Example agent response
Plan First
I will not edit any file until the plan below is acknowledged.
Impact Map
- Backend: contract + behavior. Add `analyticsEnabled` to the organization settings endpoint and an authorization rule that exposes it only to org admins.
- Frontend (web): state. Read the toggle and conditionally render the analytics section in the admin panel.
- AI Pipeline: behavior. The nightly summarization job reads the toggle and skips work for orgs with it disabled.
- Infrastructure: config. Promote the new setting key in the environment configuration template.
Dependencies
- Organization Service -> Admin Panel : contract
- Organization Service -> Summarization Worker : behavior
- Infrastructure Module -> Organization Service : config
Invariants Applied
- Authorization preserved: only org admins can read or write the toggle.
- Tenant isolation preserved: setting scoped to the owning organization.
- Contract backward compatible: new field is optional with a safe default.
- Configuration externalized: no hardcoded environment values.
Execution Strategy
- contract-first
- Rationale: consumers must be able to read the setting before they branch on it.
Implementation Order
1. Backend: add field, migration, authorization rule, contract test.
2. Web: read field, conditional render, unit test.
3. Pipeline: branch on field, skip path test.
4. Infra: promote key, validate template.
Gate Results
- Gate A: pass - contract test confirms new optional field.
- Gate B: pass - role check covered, tenant scope unchanged.
- Gate C: pass - migration adds nullable column with a default.
- Gate D: pass - IaC fmt/validate clean; deployment order documented.
- Gate E: pass - new logs do not include sensitive identifiers.
- Gate F: pass - plan-first applied; single sequencing strategy chosen.
Risks and Next Actions
- Risk: pipeline worker rollout must follow backend rollout.
- Next: schedule removal of the legacy unconditional path after the next release.
By purchasing this skill, you agree to our terms of service
CLAUDE-4-7-OPUS
AI agents make conflicting decisions when one task touches backend, frontend, infra, and AI pipelines. This is the Cursor-native self-contained skill: drop the .mdc into .cursor/rules/ and the Agent enforces impact mapping, locked invariants, sequencing strategy, and 6 verification gates. Includes skill identifier, when-to-use rules, setup instructions, trigger conditions, forbidden behaviors, two worked examples, and a strict response contract. A matching Claude Code edition is sold separately.
...more
Added 3 weeks ago
