@skillrecordings/cli 0.10.2 → 0.11.1

package/dist/{pipeline-DRKFIH73.js → pipeline-P3CLQVOL.js}

@@ -53,7 +53,7 @@ import {
   storeDraftSuccess,
   validate,
   validateSync
-} from "./chunk-TFLLCHUX.js";
+} from "./chunk-JKFLTZHH.js";
 import "./chunk-KEV3QKXP.js";
 import "./chunk-HK3PEWFD.js";
 import "./chunk-WYKL32C3.js";
@@ -117,4 +117,4 @@ export {
   validate,
   validateSync
 };
-//# sourceMappingURL=pipeline-DRKFIH73.js.map
+//# sourceMappingURL=pipeline-P3CLQVOL.js.map

package/dist/src-IKPHU6BF.js

@@ -0,0 +1,52 @@
+import {
+  ActionsTable,
+  ApprovalRequestsTable,
+  AppsTable,
+  AuditLogTable,
+  ConversationsTable,
+  DeadLetterQueueTable,
+  IdempotencyKeysTable,
+  TrustScoresTable,
+  WebhookPayloadSnapshotsTable,
+  and,
+  asc,
+  closeDb,
+  database,
+  desc,
+  env,
+  eq,
+  getDb,
+  gt,
+  gte,
+  lt,
+  lte,
+  or,
+  sql
+} from "./chunk-TBO2ZHJH.js";
+import "./chunk-WFANXVQG.js";
+export {
+  ActionsTable,
+  ApprovalRequestsTable,
+  AppsTable,
+  AuditLogTable,
+  ConversationsTable,
+  DeadLetterQueueTable,
+  IdempotencyKeysTable,
+  TrustScoresTable,
+  WebhookPayloadSnapshotsTable,
+  and,
+  asc,
+  closeDb,
+  database,
+  desc,
+  env,
+  eq,
+  getDb,
+  gt,
+  gte,
+  lt,
+  lte,
+  or,
+  sql
+};
+//# sourceMappingURL=src-IKPHU6BF.js.map

package/dist/src-IKPHU6BF.js.map

@@ -0,0 +1 @@
+{"version":3,"sources":[],"sourcesContent":[],"mappings":"","names":[]}

package/package.json

@@ -1,16 +1,18 @@
 {
   "name": "@skillrecordings/cli",
-  "version": "0.10.2",
+  "version": "0.11.1",
   "type": "module",
   "bin": {
     "skill": "./bin/skill.mjs"
   },
-  "files": ["dist", "bin", ".env.encrypted"],
+  "files": ["dist", "bin", "plugin"],
   "scripts": {
     "dev": "bun src/index.ts",
     "build": "tsup",
+    "build:compile": "bun build.ts",
     "check-types": "tsc --noEmit",
-    "test": "vitest --run"
+    "test": "vitest --run",
+    "plugin:sync": "bun src/index.ts plugin sync"
   },
   "devDependencies": {
     "@repo/typescript-config": "*",
@@ -18,18 +20,18 @@
     "@skillrecordings/database": "workspace:*",
     "@skillrecordings/front-sdk": "workspace:*",
     "@skillrecordings/memory": "workspace:*",
-    "@skillrecordings/sdk": "workspace:*",
     "@types/node": "^22.15.3",
     "tsup": "^8.5.0",
-    "typescript": "5.9.2"
+    "typescript": "5.9.2",
+    "@vitest/coverage-v8": "^3.0.0"
   },
   "dependencies": {
+    "@1password/sdk": "^0.2.1",
     "@axiomhq/js": "^1.3.1",
     "@inquirer/prompts": "^8.2.0",
     "age-encryption": "^0.3.0",
     "ai": "^6.0.49",
     "commander": "^12.1.0",
-    "dotenv-flow": "^4.1.0",
     "glob": "^13.0.0",
     "gray-matter": "^4.0.3",
     "mysql2": "^3.16.1",

package/plugin/.claude-plugin/plugin.json

@@ -0,0 +1,9 @@
+{
+  "name": "skill-cli",
+  "version": "1.0.0",
+  "displayName": "Skill Recordings CLI",
+  "description": "Front inbox management and support agent tooling",
+  "author": "Skill Recordings",
+  "skills": ["front-inbox"],
+  "cliVersion": "${PACKAGE_JSON_VERSION}"
+}

package/plugin/skills/front-inbox/SKILL.md

@@ -0,0 +1,360 @@
+# Front Inbox Management
+
+A Claude Code skill for managing Skill Recordings Front inboxes via the `skill` CLI.
+This skill is agent-first and optimized for rapid, safe inbox operations.
+It includes product inbox aliases, HATEOAS chaining guidance, and daily briefing workflows.
+
+## Activation
+
+Use this skill automatically when a user mentions any of the following:
+- inbox
+- front
+- triage
+- archive
+- tags
+- any product name listed in the alias table
+
+## Scope
+
+This skill is intentionally narrow.
+It focuses on Front inbox management and daily cross-inbox briefings.
+Do not use it for billing, refunds, or content support.
+
+## Required Environment
+
+- `FRONT_API_TOKEN`
+
+If the token is missing, report the issue and suggest checking `FRONT_API_TOKEN`.
+
+## Inbox Aliases
+
+These aliases resolve to Front inbox IDs.
+All aliases are case-insensitive.
+
+| Product | Inbox ID | Aliases |
+| --- | --- | --- |
+| Total TypeScript | `inb_3srbb` | `TT`, `typescript`, `total` |
+| Pro Tailwind | `inb_3pqh3` | `tailwind`, `ptw` |
+| AI Hero | `inb_4bj7r` | `ai-hero`, `aihero` |
+| Epic React | `inb_1bwzr` | `react`, `kcd` |
+| Epic Web | `inb_jqs2t` | `epicweb` |
+| Epic AI | `inb_jqs11` | `epicai` |
+| egghead | `inb_1c77r` | `egg` |
+| Just JavaScript | `inb_2odqf` | `justjs`, `jj` |
+| Testing Accessibility | `inb_3bkef` | `a11y` |
+| Pro Next.js | `inb_43olj` | `nextjs`, `pro-next` |
+
+When a user provides an alias, expand it to the corresponding inbox ID.
+If the alias is ambiguous or unknown, ask a clarifying question.
+
+## Command Reference
+
+Always prefer JSON output for agent parsing.
+Use `--json` when running commands.
+
+### `skill front inbox`
+
+Purpose: list all inboxes with pending counts.
+
+Examples:
+```bash
+skill front inbox --json
+```
+
+### `skill front triage`
+
+Purpose: get pending conversations in JSON for a specific inbox.
+
+Examples:
+```bash
+skill front triage -i inb_3srbb --json
+```
+
+### `skill front conversation`
+
+Purpose: fetch full conversation with messages.
+
+Examples:
+```bash
+skill front conversation cnv_123 -m --json
+```
+
+### `skill front bulk-archive`
+
+Purpose: preview bulk actions with `--dry-run`.
+
+Examples:
+```bash
+skill front bulk-archive -i inb_3srbb --filter "tag:handled" --dry-run --json
+```
+
+### `skill front tags`
+
+Purpose: list all tags used in an inbox.
+
+Examples:
+```bash
+skill front tags inb_3srbb --json
+```
+
+## HATEOAS Chaining Rules
+
+The CLI returns HATEOAS metadata for Front resources.
+Follow these rules when chaining calls.
+
+Rules:
+- Always use `--json` for agent parsing.
+- Follow `_actions` from responses to chain operations.
+- Non-destructive actions such as `mark-read` or `tag` can be auto-executed.
+- Destructive actions such as `archive` or `delete` require explicit user approval.
+- Always run with `--dry-run` before `bulk-archive`.
+- When `_actions` has multiple items, ask the user which to prioritize.
+
+HATEOAS response fields:
+- `_type` identifies the entity type.
+- `_links` includes related resources.
+- `_actions` includes next-step operations.
+
+## Daily Briefing Flow
+
+Goal: produce a cross-inbox summary and pick a focus area.
+
+Steps:
+1. Query all inboxes in parallel with `skill front triage`.
+2. Extract pending counts and urgent tags.
+3. Synthesize a concise summary.
+4. Ask the user: “What should I focus on?”
+5. Execute actions based on the user’s response.
+
+Example summary format:
+- `TT: 5 pending (2 urgent), AI Hero: 3 pending (0 urgent)`
+
+## Error Handling
+
+Connection errors:
+- Suggest checking `FRONT_API_TOKEN`.
+
+Rate limits:
+- Use exponential backoff.
+- Retry with increasing delays before giving up.
+
+Malformed responses:
+- Fall back to text output and ask the user how to proceed.
+
+## Working Notes
+
+Prefer to work inbox-by-inbox unless the user explicitly asks for a global sweep.
+Confirm the inbox ID before destructive operations.
+Respect human-in-the-loop approval requirements.
+
+## Examples
+
+See the examples in `examples/` for real command flows.
+Use these as templates and adapt to the user’s needs.
+
+---
+
+## Extended Guidance
+
+The following sections provide more detailed operational guidance.
+They are intentionally verbose to support consistent behavior.
+
+### Triage Principles
+
+Triage is for sorting actionable vs non-actionable conversations.
+Use tags to mark status, urgency, and next steps.
+Always log actions in a short summary.
+
+Common triage categories:
+- Actionable
+- Needs follow-up
+- Duplicate
+- Noise
+- Spam
+
+### Tagging Standards
+
+Use tags consistently across inboxes.
+If a tag does not exist, ask before creating a new one.
+Avoid creating near-duplicate tags.
+
+Examples of standard tags:
+- `urgent`
+- `billing`
+- `account`
+- `bug`
+- `feature-request`
+
+### Conversation Review
+
+When reading a conversation:
+- Check the subject line.
+- Check sender and recipient.
+- Read the latest message first.
+- Verify tags and assignee.
+
+If a response is needed:
+- Draft a short response summary.
+- Ask for approval before sending.
+
+### Bulk Archive Guidance
+
+Bulk archive should be used carefully.
+Always run with `--dry-run` first.
+Ensure the filter is specific and includes a tag or status.
+Confirm with the user before executing without `--dry-run`.
+
+### HATEOAS Examples
+
+Example: chain from inbox list to triage.
+
+1. Run inbox list.
+```bash
+skill front inbox --json
+```
+
+2. Choose the inbox with highest pending count.
+3. Follow `_actions.triage` for that inbox.
+4. Run triage with the resolved inbox ID.
+
+Example: mark read.
+
+1. Get conversation details.
+```bash
+skill front conversation cnv_123 -m --json
+```
+
+2. Follow `_actions.mark-read` and execute.
+
+Example: archive.
+
+1. Get conversation details.
+2. If `_actions.archive` exists, ask the user for approval.
+3. Execute the archive action after approval.
+
+### Daily Briefing Deep Dive
+
+The daily briefing is a repeatable flow.
+It should be fast, consistent, and non-destructive.
+
+Recommended flow order:
+- Total TypeScript
+- Pro Tailwind
+- AI Hero
+- Epic React
+- Epic Web
+- Epic AI
+- egghead
+- Just JavaScript
+- Testing Accessibility
+- Pro Next.js
+
+For each inbox:
+- Capture pending count.
+- Count urgent items.
+- Note any blocked or escalated threads.
+
+Synthesize into a single sentence summary.
+Ask the user what to focus on next.
+
+### Structured Summary Template
+
+Use this format when reporting to the user:
+
+```
+Summary: TT: X pending (Y urgent), AI Hero: X pending (Y urgent), Epic React: X pending (Y urgent)
+Recommendation: Focus on TT and AI Hero first.
+Question: What should I focus on?
+```
+
+### Suggested Next Actions
+
+If the user asks for a plan, propose one of these:
+- Start with highest urgent count.
+- Clear handled-tagged conversations via bulk archive.
+- Review recent unassigned threads.
+
+### Safety Checks
+
+Before destructive actions:
+- Confirm the inbox and filter.
+- Re-run `--dry-run` if the filter changed.
+- Ask for explicit approval.
+
+### Troubleshooting
+
+If commands fail:
+- Confirm CLI is installed and `skill` resolves.
+- Confirm `FRONT_API_TOKEN` is set.
+- Retry with `--json` for full response detail.
+
+### Escalation Path
+
+If the user reports unexpected failures:
+- Suggest running `skill front inbox --json`.
+- Ask for the error output.
+- Provide targeted guidance based on the error.
+
+### Support Boundaries
+
+Do not send messages directly.
+Always ask for approval before drafting or sending replies.
+Avoid auto-archiving when context is unclear.
+
+### Recap Checklist
+
+Use this checklist before closing a triage session:
+- Inbox processed
+- Urgent items identified
+- Bulk actions reviewed with `--dry-run`
+- Summary shared with the user
+
+### Additional References
+
+- `skill front inbox`
+- `skill front triage`
+- `skill front conversation`
+- `skill front bulk-archive`
+- `skill front tags`
+
+### Output Format Expectations
+
+JSON output should be parseable and consistent.
+If JSON output fails, fall back to text and ask for guidance.
+
+### Example: Alias Expansion
+
+User: “Check TT inbox.”
+Action: Expand `TT` to `inb_3srbb` and proceed with triage.
+
+User: “Check Pro Next.”
+Action: Expand `pro-next` to `inb_43olj` and proceed.
+
+### Example: Urgent Tag Sweep
+
+Run triage for each inbox.
+Collect tags with `urgent` or `escalated`.
+Report counts and ask for focus.
+
+### Example: Safe Bulk Archive
+
+1. Run `--dry-run` with a specific tag filter.
+2. Share the preview with the user.
+3. If approved, re-run without `--dry-run`.
+
+### Example: Conversation Detail Follow-up
+
+When the user asks for details:
+- Fetch the conversation.
+- Summarize the latest message.
+- Offer next actions using `_actions`.
+
+### Example: Cross-Inbox Summary
+
+Provide a compact summary in one line.
+Follow it with a clear question asking for focus.
+
+### End State
+
+This skill should make inbox management predictable and safe.
+Always prefer clarity over speed when making decisions.

package/plugin/skills/front-inbox/examples/bulk-archive.md

@@ -0,0 +1,16 @@
+# Example: Bulk Archive with Dry Run
+
+Goal: preview handled-tag conversations before archiving.
+
+Dry run:
+```bash
+skill front bulk-archive -i inb_3srbb --filter "tag:handled" --dry-run --json
+```
+
+Review the preview output with the user.
+Only proceed after explicit approval.
+
+Execute after approval:
+```bash
+skill front bulk-archive -i inb_3srbb --filter "tag:handled" --json
+```

package/plugin/skills/front-inbox/examples/triage-tt.md

@@ -0,0 +1,16 @@
+# Example: Triage Total TypeScript
+
+Goal: fetch pending conversations for Total TypeScript and summarize urgency.
+
+Command:
+```bash
+skill front triage -i inb_3srbb --json
+```
+
+Expected flow:
+- Parse JSON response.
+- Note pending count and urgent tags.
+- Ask the user what to focus on next.
+
+Follow-up question:
+What should I focus on in the TT inbox?