- Backend: FastAPI app with all models, services, schemas, routes - Admin Panel: Next.js dashboard with all modules (CRM, POS, repairs, warranties, billing, notifications, etc.) - Ecommerce: storefront (submodule) + product catalog - AI Agents: .agents/ skill library for Antigravity IDE - Docs: architecture, billing, notification gateway markdown - Submodules: Backend/OpenWA, Ecommerce/File 2/storefront
13 KiB
13 KiB
iFixKart Platform: Custom Agents & Skills Directory
This document details the configuration structure of the .agents customization framework and provides a comprehensive breakdown of the 21 custom skills configured for this workspace.
📂 Part 1: Workspace Customization Structure
The workspace customizations folder is located in the root of the project at IfixKartEcommerce/.agents/. It contains:
- AGENTS.md: The master configuration file. It contains behavioral prompt reinforcements (directives instructing the assistant to automatically discover, load, and strictly execute skill checks) and lists all registered skills.
- skills/: The subfolder containing individual skill directories. Each skill is separated into its own folder (e.g.
skills/create_component/) containing aSKILL.mdfile.
Each SKILL.md declares:
- YAML Frontmatter: The
nameanddescriptiontriggers matched during task parsing. - Purpose: The core objective of the skill.
- When to Use: Scenarios where the skill applies.
- Preconditions: Technical prerequisites.
- Step-by-Step Checklist: How to execute the work.
- Validation Checklist: Verification criteria before task completion.
- Examples: Real codebase code snippets.
- Failure Cases: Known anti-patterns to avoid.
🛠️ Part 2: Complete Skills Catalog
The 21 active workspace skills are organized into four operational categories: Backend (Python/FastAPI), Frontend (Next.js/React), Generic (Lint/Review/Audit), and Standard (Version Control/CI).
📦 Category A: Backend Skills
16. Database Migrations (.agents/skills/database_migrations/)
- Purpose: Manage database schema migrations using Alembic in Python.
- Preconditions: alembic.ini configured and pointing to database instances.
- Key Checklist: Generate autodetected revision scripts; verify migration logic; apply upgrade heads locally.
- Validation: Alembic schema version aligns with database version; zero data truncation logs.
- Failure Cases: Making manual direct DDL edits in MySQL, bypassing versioning loops.
17. API Design (.agents/skills/api_design/)
- Purpose: Enforce RESTful routing paths, payloads validation, and response models.
- Preconditions: Pydantic/Zod schemas declared.
- Key Checklist: Structure route paths logically; match correct HTTP request verbs; assert standard status responses.
- Validation: Swagger UI loads route mappings correctly.
- Failure Cases: Using GET endpoints to perform data modifications.
1. Add Tests (.agents/skills/add_tests/)
- Purpose: Writes unit, integration, and regression test suites for FastAPI backend routes and database models.
- Preconditions: Local MySQL testing connection must be configured; virtualenv testing packages must be active.
- Key Checklist: Mock dynamic payment and shipping gateways; write isolated assertions covering happy paths and exception states.
- Validation: Code has at least 85% assertion coverage; tests clean up database changes upon completion.
- Failure Cases: Mocking database operations incorrectly, leading to shared state leaks between test threads.
2. Code Review (.agents/skills/code_review/)
- Purpose: Inspects backend logic, SQL transaction blocks, and route permissions.
- Preconditions: Code compiles successfully; Graphify model is generated.
- Key Checklist: Check database mutations for session rollbacks on exceptions; scan routes to ensure auth decorators are in place.
- Validation: Zero syntax or compile warnings; all inputs validated via Pydantic schemas.
- Failure Cases: Failing to verify that database mutations are enclosed in an active session rollback block.
3. Fix Bug (.agents/skills/fix_bug/)
- Purpose: Isolates database, schema, or router failures, fixes the root cause, and guarantees regression protection.
- Preconditions: Traceback logs or bug descriptions available.
- Key Checklist: Write a minimal failing test that reproduces the bug; apply logic fixes to the root cause; verify the test now passes.
- Validation: Reproduction test succeeds; existing suite continues to pass with zero regressions.
- Failure Cases: Adding conditional checks that temporarily hide symptoms instead of fixing the root cause.
🎨 Category B: Frontend Skills
4. Add Utility Function (.agents/skills/add_utility_function/)
- Purpose: Adds pure, reusable TypeScript helper functions.
- Preconditions: Similar helper does not already exist under
utils/orlib/. - Key Checklist: Write helper as a pure, side-effect-free function; type all inputs and outputs (no
anytype); export from barrelindex.ts. - Validation: Linter outputs zero errors; all parameter configurations documented with JSDoc headers.
- Failure Cases: Importing React state context or hooks inside pure utility files.
5. Create Component (.agents/skills/create_component/)
- Purpose: Enforces standard layouts for Next.js/React visual modules.
- Preconditions: Verify no primitive layouts (like Shadcn or AntD) already meet the design requirement.
- Key Checklist: Group files under
components/ComponentName/; define destructured TS interface props; style only with design token variables (no inline styling or raw color hex codes). - Validation: Responsive layout renders accurately on mobile and desktop viewports; barrel export index is created.
- Failure Cases: Bundling layout display elements and network fetching/state mutations into a single massive file.
6. Test Feature (.agents/skills/test_feature/)
- Purpose: Evaluates user interface responsiveness and checkout/booking validation flows.
- Preconditions: Local dev servers are running.
- Key Checklist: Trigger browser subagent to open local url; interact with input elements; capture responsive screenshots and session recordings.
- Validation: Form submissions yield confirmation screens; console shows zero javascript warnings.
- Failure Cases: Selecting elements via absolute pixel coords instead of relative text or ID labels.
🛡️ Category C: Generic Skills
21. Planning and Analysis (.agents/skills/planning_analysis/)
- Purpose: Enforces requirements check, trade-off analysis (pros/cons), detailed report generation, and explicit user approval before execution.
- Preconditions: Staging environment and Graphify maps are accessible.
- Key Checklist: Deconstruct task constraints; compare two alternative strategies; list pros/cons; output
implementation_plan.mdwith feedback hooks; wait for approval. - Validation: Code remains unmodified until user explicitly reviews and accepts the implementation plan.
- Failure Cases: Commencing execution immediately without analyzing disadvantages or getting approval.
18. Performance Profiling (.agents/skills/performance_profiling/)
- Purpose: Identify database index gaps, slow SQL executions, or CPU-bound script blocks.
- Preconditions: Profiling libraries active.
- Key Checklist: Log execution metrics; inspect queries using EXPLAIN; implement eager-load configurations or add database indexes.
- Validation: Target API latencies drop below 200ms.
- Failure Cases: Injecting indices indiscriminately to tables, increasing write overhead latencies.
19. Logging Observability (.agents/skills/logging_observability/)
- Purpose: Enforces structured JSON logging and exception capture metrics.
- Preconditions: JSON logging provider active.
- Key Checklist: Instantiate standard logger objects; log key actions; propagate correlation request IDs.
- Validation: Log pipelines capture metadata and traceback details.
- Failure Cases: Logging plaintext personal details (PII) or secrets.
7. Add Docs (.agents/skills/add_docs/)
- Purpose: Enforces documentation structures, API parameter documentation, and code comment standards.
- Preconditions: Exposing new methods or modules.
- Key Checklist: Write JSDoc comments explaining class objectives, parameter configurations, and return formats; organize imports (React
\rightarrowlibraries\rightarrowlocal files). - Validation: All exported interfaces are documented; compiler yields no warnings.
- Failure Cases: Letting JSDoc variables fall out of sync with actual parameters after code revisions.
8. Review My Pr (.agents/skills/review_my_pr/)
- Purpose: Runs a self-audit before PR submission.
- Preconditions: Branch builds correctly; git diff contains staged changes.
- Key Checklist: Execute linters; inspect route declarations for auth tokens; check styling rules; compile review report.
- Validation: Zero linter errors found; no credentials or private keys committed.
- Failure Cases: Merging PR files without verifying integration points.
9. Review (.agents/skills/review/)
- Purpose: Runs deep system-wide quality reviews, assessing performance bottlenecks and business-requirement fit.
- Preconditions: Complete test suite passes.
- Key Checklist: Map code changes using Graphify; audit for performance bottlenecks (loops, database indices); verify design guidelines.
- Validation: Review log lists all potential transaction issues.
- Failure Cases: Auditing logic without checking performance implications (e.g. ignoring unindexed query lookups).
10. Security Review (.agents/skills/security_review/)
- Purpose: Scans codebases for authentication bypasses, database injection vulnerabilities, XSS, and data privacy leaks.
- Preconditions: Database schemas and routers are declared.
- Key Checklist: Ensure input values are parameterized in DB calls; check route handlers require auth tokens; sanitize strings before page injections.
- Validation: SQL injections are prevented via ORM bindings; user data is handled under DPDP act parameters.
- Failure Cases: Trusting third-party input streams without cryptographically checking token signatures.
11. Update Graphify (.agents/skills/update_graphify/)
- Purpose: Refreshes Graphify AST model linkages after folder structure modifications.
- Preconditions: Graphify package installed.
- Key Checklist: Run update script targeting modified files; check parser outputs; refresh HTML graphs and markdown wiki files.
- Validation: Node list contains the new connections.
- Failure Cases: Running graph builders on code files containing syntax compile errors.
12. Graphify (.agents/skills/graphify/)
- Purpose: Builds an AST knowledge graph mapping visual assets, database relationships, and backend APIs.
- Preconditions: Python environment configured.
- Key Checklist: Execute graph parser; generate
GRAPH_REPORT.mdoutlining codebase components; output structure maps. - Validation: Interactive system layout is accessible via
graph.html. - Failure Cases: Missing write permissions inside Documentation folder.
13. GBrain (.agents/skills/gbrain/)
- Purpose: Connects workspace memory and guidelines directly to coding assistants.
- Preconditions: Bun package runner installed.
- Key Checklist: Ingest data logs using capture CLI; retrieve cited summaries using think commands; run background dream cycles.
- Validation: Output answers show clear, cited links to raw pages.
- Failure Cases: Querying index lists containing stale/out-of-date records.
⚙️ Category D: Standard Skills
20. Dependency Updates (.agents/skills/dependency_updates/)
- Purpose: Automate package audits, vulnerability fixes, and minor upgrades.
- Preconditions: Access to standard package lock files.
- Key Checklist: Scan for alert vulnerabilities; upgrade minor version tags; run regression tests.
- Validation: Diagnostic audits return zero active warnings.
- Failure Cases: Updating major package indices blindly, causing system compile failures.
14. Commit (.agents/skills/commit/)
- Purpose: Standardizes version control logs using Conventional Commits structure.
- Preconditions: Staged file modifications are ready in Git.
- Key Checklist: Inspect staged changes; categorize subject lines (feat, fix, docs, refactor); document why changes were made.
- Validation: Log headers match
<type>(<scope>): <desc>syntax with no capitalizations. - Failure Cases: Combining unrelated UI tweaks and database models under a single generic commit title.
15. Update Agents (.agents/skills/update_agents/)
- Purpose: Synchronizes customization setup rules across Devin and Kiro environments.
- Preconditions: Devin and Kiro configuration directories are active.
- Key Checklist: Check modified rule files; align AGENTS.md instructions with matching steering parameters; isolate workspace readme layouts.
- Validation: Setup structures are updated without duplication.
- Failure Cases: Wholesale overwriting of configurations, which breaks platform-specific hooks.