context-router_skill

This skill routes tasks through authoritative documentation before code changes to prevent regressions and ensure correct context.
  • JavaScript

0

GitHub Stars

3

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill splitleaseteam/splitlease_monorepo --skill context-router

  • DOMAIN_INDEX.md4.5 KB
  • REGRESSION_GUARD.md3.6 KB
  • SKILL.md3.3 KB

Overview

This skill routes tasks to authoritative documentation before making code changes to ensure correct context is loaded and regressions are avoided. It enforces a pre-change checklist so planning always references the right page, backend, database, and auth docs. Use it to reduce context drift and prevent reintroducing past bugs.

How this skill works

When a task will modify code, the skill analyzes the requested change to identify affected domains (pages, backend/Edge functions, database, auth, routing). It then loads the mapped authoritative documents from the domain index and checks for existing regression guard tags in git. Only after confirming the relevant documentation and any guard constraints does the planner proceed, and it records which docs were consulted.

When to use it

  • Before implementing any change that touches UI pages, backend functions, or database schemas
  • When debugging issues to ensure historical fixes and constraints are considered
  • During planning phase of a multi-file or cross-cutting change
  • Before altering authentication or routing logic
  • When a task may reintroduce known regressions or requires specific guard patterns

Best practices

  • Always map the task to affected domains first (page, backend, db, auth, routing)
  • Load the exact domain docs listed in the domain index for each affected area
  • Query git for guard/* tags to discover prior fixes and mandatory patterns
  • Record the consulted documents in the plan to keep the change audit-friendly
  • Apply this skill in Phase 2 (Planning) so execution has full, authoritative context

Example use cases

  • Fixing a submission flow on a dashboard page: load page, proposal backend, and proposal DB docs
  • Adding a new route that interacts with auth: load routing, auth, and related page docs
  • Changing a database schema: load DATABASE_*.md plus any affected page and backend docs
  • Debugging intermittent production behavior: load domain docs and regression guard tags before proposing fixes
  • Planning a cross-cutting refactor: use largeCLAUDE.md plus all affected domain docs

FAQ

Flag the missing entry, proceed with caution, and escalate to owners to add the authoritative doc before major changes.

When should I check git guard tags?

Always run the guard/* tag query during planning for changes that touch persisted state, critical flows, or recurring bugs.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational