diff --git a/README.md b/README.md
index bba9451fb..b294e2304 100644
--- a/README.md
+++ b/README.md
@@ -82,18 +82,18 @@ Captured from this machine to document the local development environment:
 
 ## Developer Persona
 
-- **New contributors** follow the 2‑3 hour path in [`docs/getting-started/NEW_CONTRIBUTOR_PATH.md`](./docs/getting-started/NEW_CONTRIBUTOR_PATH.md), which guides them through overview → architecture → testing → practical tasks to build confidence quickly.
+- **New contributors** follow the 2‑3 hour path in [docs/getting-started/NEW_CONTRIBUTOR_PATH.md](./docs/getting-started/NEW_CONTRIBUTOR_PATH.md), which guides them through overview → architecture → testing → practical tasks to build confidence quickly.
 - **Core mindset**: favor declarative JSON/Lua configuration, filter every query by `tenantId`, keep components under 150 LOC, and rely on `RenderComponent` plus Prisma helpers instead of hardcoded UI/data layer logic.
 - **Roles**: dashboards/authn (Level 2), admin control panels (Level 3), builder/IDE workflows (Level 4), and SuperGod tenant operations (Level 5) are the main focus areas when shaping experiences for developers, architects, and operators.
 
 ## TODO List
 
-- The curated checklist at [`docs/todo/README.md`](./docs/todo/README.md) lists every area (architecture, frontend, DBAL, packages, security, deployment, docs, tooling, etc.) with numbered TODO files ordered by priority.
+- The curated checklist at [docs/todo/README.md](./docs/todo/README.md) lists every area (architecture, frontend, DBAL, packages, security, deployment, docs, tooling, etc.) with numbered TODO files ordered by priority.
 - Pick the list that matches your work, mark tasks with `[x]`, and consult the generated reports in `docs/todo/TODO_STATUS.md` and `docs/todo/22-TODO-SCAN.md` for repo-wide tracking.
 
 ## AI Prompts
 
-- `.github/prompts/` stores curated prompts for planning, designing, implementing, testing, reviewing, deploying, and maintaining features; [`docs/todo/core/21-SDLC-TODO.md`](./docs/todo/core/21-SDLC-TODO.md) describes which prompts need review or updates.
+- `.github/prompts/` stores curated prompts for planning, designing, implementing, testing, reviewing, deploying, and maintaining features; [docs/todo/core/21-SDLC-TODO.md](./docs/todo/core/21-SDLC-TODO.md) describes which prompts need review or updates.
 - Keep `.github/copilot-instructions.md` aligned with architecture/docs changes so Copilot-based automation and review prompts stay accurate.
 
 ## Refactor Plan Snapshot
@@ -893,7 +893,7 @@ npm run act:build     # Test build only
 npm run act:diagnose  # Check setup (no Docker)
 ```
 
-Refer to [`docs/guides/ACT_CHEAT_SHEET.md`](./docs/guides/ACT_CHEAT_SHEET.md) or the detailed [`docs/guides/ACT_TESTING.md`](./docs/guides/ACT_TESTING.md) guide to mirror GitHub Actions workflows locally and understand each Act mode.
+Refer to [docs/guides/ACT_CHEAT_SHEET.md](./docs/guides/ACT_CHEAT_SHEET.md) or the detailed [docs/guides/ACT_TESTING.md](./docs/guides/ACT_TESTING.md) guide to mirror GitHub Actions workflows locally and understand each Act mode.
 
 ---