@hasna/instructions 0.4.5 → 0.4.6

package/README.md

@@ -189,8 +189,9 @@ before mutating managed files.
 `instructions init` and `bun run seed` also seed
 `global-agent-rules-standard`, the managed global/system prompt source for
 session renaming, task-scoped worktrees, PR-first landing, protected-branch
-push safety, autonomous repair, Hasna CLI source-of-truth usage, conversation
-surface routing, and unbudgeted Codewith goals unless a user asks for budgets.
+push safety, no brittle hardcoding when source-of-truth or reusable abstractions
+exist, autonomous repair, Hasna CLI source-of-truth usage, conversation surface
+routing, and unbudgeted Codewith goals unless a user asks for budgets.
 
 ## Machine-aware Profiles
 

package/dist/cli/index.js

@@ -6861,6 +6861,7 @@ async function ensurePlatformProfiles(store = resolveConfigStore()) {
 // src/lib/global-agent-rules-standard.ts
 init_config_store();
 var GLOBAL_AGENT_RULES_STANDARD_SLUG = "global-agent-rules-standard";
+var NO_BRITTLE_HARDCODING_RULE = "Do not hardcode brittle values, paths, provider names, config, business logic, environment-specific IDs, or one-off mappings when a source-of-truth, schema/config-driven, package-owned, reusable, or cleaner abstraction exists. This is especially strict in medium and large applications. Explicit constants, fixtures, tests, and temporary compatibility shims are allowed only when scoped, named, and justified.";
 var GLOBAL_AGENT_RULES_STANDARD_CONTENT = `# Global Coding Agent Rules Standard
 
 This standard is managed global/system prompt source content for Hasna coding
@@ -6885,24 +6886,25 @@ policy source supersedes them.
    request before landing.
 5. Never push directly to \`main\`, the default branch, or any protected branch
    unless the user explicitly instructs that exact repo and operation.
+6. ${NO_BRITTLE_HARDCODING_RULE}
 
 ## Autonomy and Source-of-Truth Tools
 
-6. Act autonomously. Diagnose, repair, validate, and iterate on the owning
+7. Act autonomously. Diagnose, repair, validate, and iterate on the owning
    CLIs, packages, and workflows before asking the user. Ask only when blocked
    by destructive decisions, secret-bearing decisions, user-only authority, or
    external state the agent cannot safely obtain.
-7. Use Hasna CLIs/packages as the source of truth: \`todos\`, \`conversations\`,
+8. Use Hasna CLIs/packages as the source of truth: \`todos\`, \`conversations\`,
    \`mementos\`, \`knowledge\`, \`projects\`, \`repos\`, \`accounts\`,
    \`instructions\`, \`machines\`, \`secrets\`, and \`access\`.
-8. Secrets safety is mandatory. Never expose secrets in prompts, tasks,
+9. Secrets safety is mandatory. Never expose secrets in prompts, tasks,
    memories, conversations, manifests, reports, logs, PR text, or any other
    agent-visible output. Reference vault item names, secret identifiers, and
    access grants only; never print credential values.
 
 ## Conversation Surfaces
 
-9. Use default conversation surfaces correctly: \`announcements\` for policy,
+10. Use default conversation surfaces correctly: \`announcements\` for policy,
    freeze, breaking, cutover, and release notices; \`incidents\` for outages,
    crash loops, data risk, or security exposure; \`git-publishing\` before and
    after package publishes; \`git-prs\`, \`git-commits\`, and \`git-releases\`

package/dist/index.js

@@ -3213,6 +3213,7 @@ async function ensurePlatformProfiles(store = resolveConfigStore()) {
 }
 // src/lib/global-agent-rules-standard.ts
 var GLOBAL_AGENT_RULES_STANDARD_SLUG = "global-agent-rules-standard";
+var NO_BRITTLE_HARDCODING_RULE = "Do not hardcode brittle values, paths, provider names, config, business logic, environment-specific IDs, or one-off mappings when a source-of-truth, schema/config-driven, package-owned, reusable, or cleaner abstraction exists. This is especially strict in medium and large applications. Explicit constants, fixtures, tests, and temporary compatibility shims are allowed only when scoped, named, and justified.";
 var GLOBAL_AGENT_RULES_STANDARD_CONTENT = `# Global Coding Agent Rules Standard
 
 This standard is managed global/system prompt source content for Hasna coding
@@ -3237,24 +3238,25 @@ policy source supersedes them.
    request before landing.
 5. Never push directly to \`main\`, the default branch, or any protected branch
    unless the user explicitly instructs that exact repo and operation.
+6. ${NO_BRITTLE_HARDCODING_RULE}
 
 ## Autonomy and Source-of-Truth Tools
 
-6. Act autonomously. Diagnose, repair, validate, and iterate on the owning
+7. Act autonomously. Diagnose, repair, validate, and iterate on the owning
    CLIs, packages, and workflows before asking the user. Ask only when blocked
    by destructive decisions, secret-bearing decisions, user-only authority, or
    external state the agent cannot safely obtain.
-7. Use Hasna CLIs/packages as the source of truth: \`todos\`, \`conversations\`,
+8. Use Hasna CLIs/packages as the source of truth: \`todos\`, \`conversations\`,
    \`mementos\`, \`knowledge\`, \`projects\`, \`repos\`, \`accounts\`,
    \`instructions\`, \`machines\`, \`secrets\`, and \`access\`.
-8. Secrets safety is mandatory. Never expose secrets in prompts, tasks,
+9. Secrets safety is mandatory. Never expose secrets in prompts, tasks,
    memories, conversations, manifests, reports, logs, PR text, or any other
    agent-visible output. Reference vault item names, secret identifiers, and
    access grants only; never print credential values.
 
 ## Conversation Surfaces
 
-9. Use default conversation surfaces correctly: \`announcements\` for policy,
+10. Use default conversation surfaces correctly: \`announcements\` for policy,
    freeze, breaking, cutover, and release notices; \`incidents\` for outages,
    crash loops, data risk, or security exposure; \`git-publishing\` before and
    after package publishes; \`git-prs\`, \`git-commits\`, and \`git-releases\`

package/dist/lib/global-agent-rules-standard.d.ts

@@ -1,6 +1,7 @@
 import type { Config } from "../types/index.js";
 import { type ConfigStore } from "../data/config-store.js";
 export declare const GLOBAL_AGENT_RULES_STANDARD_SLUG = "global-agent-rules-standard";
-export declare const GLOBAL_AGENT_RULES_STANDARD_CONTENT = "# Global Coding Agent Rules Standard\n\nThis standard is managed global/system prompt source content for Hasna coding\nagents. Rendered agents must receive these rules unless a newer authorized\npolicy source supersedes them.\n\n## Session and Planning Defaults\n\n1. Use automatic session renaming when the agent supports it. Rename the\n   session early to match the active task; if the task materially pivots, rename\n   it again so operators can identify the run.\n2. Never set Codewith goal/token budgets or goal-plan budgets unless the user\n   explicitly asks for a budget. Durable goals and goal plans are unbudgeted by\n   default.\n\n## Repository Mutation and Landing\n\n3. Repo mutation must happen in a task-scoped worktree. First inspect the canonical worktree root\n   `$HOME/.hasna/repos/worktrees`; prefer Hasna repo/project worktree\n   mechanisms when available; otherwise use `git worktree`. Never mutate shared checkouts.\n4. PR-first landing: normal changes go through a branch/worktree and pull\n   request before landing.\n5. Never push directly to `main`, the default branch, or any protected branch\n   unless the user explicitly instructs that exact repo and operation.\n\n## Autonomy and Source-of-Truth Tools\n\n6. Act autonomously. Diagnose, repair, validate, and iterate on the owning\n   CLIs, packages, and workflows before asking the user. Ask only when blocked\n   by destructive decisions, secret-bearing decisions, user-only authority, or\n   external state the agent cannot safely obtain.\n7. Use Hasna CLIs/packages as the source of truth: `todos`, `conversations`,\n   `mementos`, `knowledge`, `projects`, `repos`, `accounts`,\n   `instructions`, `machines`, `secrets`, and `access`.\n8. Secrets safety is mandatory. Never expose secrets in prompts, tasks,\n   memories, conversations, manifests, reports, logs, PR text, or any other\n   agent-visible output. Reference vault item names, secret identifiers, and\n   access grants only; never print credential values.\n\n## Conversation Surfaces\n\n9. Use default conversation surfaces correctly: `announcements` for policy,\n   freeze, breaking, cutover, and release notices; `incidents` for outages,\n   crash loops, data risk, or security exposure; `git-publishing` before and\n   after package publishes; `git-prs`, `git-commits`, and `git-releases`\n   for repository landing events; `hq` for broad coordination;\n   `agent-policy` for agent operating-rule discussion; project/product\n   channels for normal work; and `conversations blockers` for blocker\n   discovery. Do not invent or refer to a literal blockers channel.\n";
+export declare const NO_BRITTLE_HARDCODING_RULE = "Do not hardcode brittle values, paths, provider names, config, business logic, environment-specific IDs, or one-off mappings when a source-of-truth, schema/config-driven, package-owned, reusable, or cleaner abstraction exists. This is especially strict in medium and large applications. Explicit constants, fixtures, tests, and temporary compatibility shims are allowed only when scoped, named, and justified.";
+export declare const GLOBAL_AGENT_RULES_STANDARD_CONTENT = "# Global Coding Agent Rules Standard\n\nThis standard is managed global/system prompt source content for Hasna coding\nagents. Rendered agents must receive these rules unless a newer authorized\npolicy source supersedes them.\n\n## Session and Planning Defaults\n\n1. Use automatic session renaming when the agent supports it. Rename the\n   session early to match the active task; if the task materially pivots, rename\n   it again so operators can identify the run.\n2. Never set Codewith goal/token budgets or goal-plan budgets unless the user\n   explicitly asks for a budget. Durable goals and goal plans are unbudgeted by\n   default.\n\n## Repository Mutation and Landing\n\n3. Repo mutation must happen in a task-scoped worktree. First inspect the canonical worktree root\n   `$HOME/.hasna/repos/worktrees`; prefer Hasna repo/project worktree\n   mechanisms when available; otherwise use `git worktree`. Never mutate shared checkouts.\n4. PR-first landing: normal changes go through a branch/worktree and pull\n   request before landing.\n5. Never push directly to `main`, the default branch, or any protected branch\n   unless the user explicitly instructs that exact repo and operation.\n6. Do not hardcode brittle values, paths, provider names, config, business logic, environment-specific IDs, or one-off mappings when a source-of-truth, schema/config-driven, package-owned, reusable, or cleaner abstraction exists. This is especially strict in medium and large applications. Explicit constants, fixtures, tests, and temporary compatibility shims are allowed only when scoped, named, and justified.\n\n## Autonomy and Source-of-Truth Tools\n\n7. Act autonomously. Diagnose, repair, validate, and iterate on the owning\n   CLIs, packages, and workflows before asking the user. Ask only when blocked\n   by destructive decisions, secret-bearing decisions, user-only authority, or\n   external state the agent cannot safely obtain.\n8. Use Hasna CLIs/packages as the source of truth: `todos`, `conversations`,\n   `mementos`, `knowledge`, `projects`, `repos`, `accounts`,\n   `instructions`, `machines`, `secrets`, and `access`.\n9. Secrets safety is mandatory. Never expose secrets in prompts, tasks,\n   memories, conversations, manifests, reports, logs, PR text, or any other\n   agent-visible output. Reference vault item names, secret identifiers, and\n   access grants only; never print credential values.\n\n## Conversation Surfaces\n\n10. Use default conversation surfaces correctly: `announcements` for policy,\n   freeze, breaking, cutover, and release notices; `incidents` for outages,\n   crash loops, data risk, or security exposure; `git-publishing` before and\n   after package publishes; `git-prs`, `git-commits`, and `git-releases`\n   for repository landing events; `hq` for broad coordination;\n   `agent-policy` for agent operating-rule discussion; project/product\n   channels for normal work; and `conversations blockers` for blocker\n   discovery. Do not invent or refer to a literal blockers channel.\n";
 export declare function ensureGlobalAgentRulesStandardConfig(store?: ConfigStore): Promise<Config>;
 //# sourceMappingURL=global-agent-rules-standard.d.ts.map

package/dist/lib/global-agent-rules-standard.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"global-agent-rules-standard.d.ts","sourceRoot":"","sources":["../../src/lib/global-agent-rules-standard.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EAAE,MAAM,EAAE,MAAM,mBAAmB,CAAC;AAChD,OAAO,EAAsB,KAAK,WAAW,EAAE,MAAM,yBAAyB,CAAC;AAE/E,eAAO,MAAM,gCAAgC,gCAAgC,CAAC;AAE9E,eAAO,MAAM,mCAAmC,ikFAiD/C,CAAC;AAEF,wBAAsB,oCAAoC,CAAC,KAAK,GAAE,WAAkC,GAAG,OAAO,CAAC,MAAM,CAAC,CA6BrH"}
+{"version":3,"file":"global-agent-rules-standard.d.ts","sourceRoot":"","sources":["../../src/lib/global-agent-rules-standard.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EAAE,MAAM,EAAE,MAAM,mBAAmB,CAAC;AAChD,OAAO,EAAsB,KAAK,WAAW,EAAE,MAAM,yBAAyB,CAAC;AAE/E,eAAO,MAAM,gCAAgC,gCAAgC,CAAC;AAE9E,eAAO,MAAM,0BAA0B,8ZAA8Z,CAAC;AAEtc,eAAO,MAAM,mCAAmC,g+FAkD/C,CAAC;AAEF,wBAAsB,oCAAoC,CAAC,KAAK,GAAE,WAAkC,GAAG,OAAO,CAAC,MAAM,CAAC,CA6BrH"}

package/dist/mcp/index.js

@@ -2113,7 +2113,7 @@ var init_sync_dir = __esm(() => {
 var require_package = __commonJS((exports, module) => {
   module.exports = {
     name: "@hasna/instructions",
-    version: "0.4.5",
+    version: "0.4.6",
     description: "AI coding agent instruction & configuration manager \u2014 store, version, apply, and share all your AI coding configs. CLI + MCP + HTTP API (instructions-serve) + generated SDK + Dashboard.",
     type: "module",
     main: "dist/index.js",

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@hasna/instructions",
-  "version": "0.4.5",
-  "description": "AI coding agent instruction & configuration manager \u2014 store, version, apply, and share all your AI coding configs. CLI + MCP + HTTP API (instructions-serve) + generated SDK + Dashboard.",
+  "version": "0.4.6",
+  "description": "AI coding agent instruction & configuration manager — store, version, apply, and share all your AI coding configs. CLI + MCP + HTTP API (instructions-serve) + generated SDK + Dashboard.",
   "type": "module",
   "main": "dist/index.js",
   "types": "dist/index.d.ts",

package/dashboard/README.md

@@ -1,73 +0,0 @@
-# React + TypeScript + Vite
-
-This template provides a minimal setup to get React working in Vite with HMR and some ESLint rules.
-
-Currently, two official plugins are available:
-
-- [@vitejs/plugin-react](https://github.com/vitejs/vite-plugin-react/blob/main/packages/plugin-react) uses [Babel](https://babeljs.io/) (or [oxc](https://oxc.rs) when used in [rolldown-vite](https://vite.dev/guide/rolldown)) for Fast Refresh
-- [@vitejs/plugin-react-swc](https://github.com/vitejs/vite-plugin-react/blob/main/packages/plugin-react-swc) uses [SWC](https://swc.rs/) for Fast Refresh
-
-## React Compiler
-
-The React Compiler is not enabled on this template because of its impact on dev & build performances. To add it, see [this documentation](https://react.dev/learn/react-compiler/installation).
-
-## Expanding the ESLint configuration
-
-If you are developing a production application, we recommend updating the configuration to enable type-aware lint rules:
-
-```js
-export default defineConfig([
-  globalIgnores(['dist']),
-  {
-    files: ['**/*.{ts,tsx}'],
-    extends: [
-      // Other configs...
-
-      // Remove tseslint.configs.recommended and replace with this
-      tseslint.configs.recommendedTypeChecked,
-      // Alternatively, use this for stricter rules
-      tseslint.configs.strictTypeChecked,
-      // Optionally, add this for stylistic rules
-      tseslint.configs.stylisticTypeChecked,
-
-      // Other configs...
-    ],
-    languageOptions: {
-      parserOptions: {
-        project: ['./tsconfig.node.json', './tsconfig.app.json'],
-        tsconfigRootDir: import.meta.dirname,
-      },
-      // other options...
-    },
-  },
-])
-```
-
-You can also install [eslint-plugin-react-x](https://github.com/Rel1cx/eslint-react/tree/main/packages/plugins/eslint-plugin-react-x) and [eslint-plugin-react-dom](https://github.com/Rel1cx/eslint-react/tree/main/packages/plugins/eslint-plugin-react-dom) for React-specific lint rules:
-
-```js
-// eslint.config.js
-import reactX from 'eslint-plugin-react-x'
-import reactDom from 'eslint-plugin-react-dom'
-
-export default defineConfig([
-  globalIgnores(['dist']),
-  {
-    files: ['**/*.{ts,tsx}'],
-    extends: [
-      // Other configs...
-      // Enable lint rules for React
-      reactX.configs['recommended-typescript'],
-      // Enable lint rules for React DOM
-      reactDom.configs.recommended,
-    ],
-    languageOptions: {
-      parserOptions: {
-        project: ['./tsconfig.node.json', './tsconfig.app.json'],
-        tsconfigRootDir: import.meta.dirname,
-      },
-      // other options...
-    },
-  },
-])
-```