portable-agent-layer 0.75.1 → 0.76.0

package/assets/agents/gemini-researcher.md

@@ -25,7 +25,7 @@ You are a research specialist focused on **depth and academic rigor**.
 
 **Always start with Gemini Search.** Use the grounded search tool for your first sub-question:
 ```bash
-bun ~/.pal/skills/research/tools/gemini-search.ts -- "<query>"
+pal cli skill run deep-research gemini-search -- "<query>"
 ```
 
 - If it returns results → **continue using Gemini Search** for remaining queries

package/assets/agents/grok-researcher.md

@@ -28,19 +28,19 @@ Use the `grok-search` tool to query the Grok API with real-time search grounding
 ### Current events / breaking news (web + X sources)
 
 ```bash
-bun ~/.pal/skills/research/tools/grok-search.ts -- "<your research query>" --sources web,x
+pal cli skill run deep-research grok-search -- "<your research query>" --sources web,x
 ```
 
 ### Social sentiment / trending topics (X only)
 
 ```bash
-bun ~/.pal/skills/research/tools/grok-search.ts -- "Search X for recent posts about: <topic>. Summarize key themes, notable accounts, and overall sentiment." --sources x
+pal cli skill run deep-research grok-search -- "Search X for recent posts about: <topic>. Summarize key themes, notable accounts, and overall sentiment." --sources x
 ```
 
 ### Web-only search
 
 ```bash
-bun ~/.pal/skills/research/tools/grok-search.ts -- "<query>" --sources web
+pal cli skill run deep-research grok-search -- "<query>" --sources web
 ```
 
 The tool outputs findings as markdown with a `## Sources` section listing URLs and X posts.

package/assets/agents/perplexity-researcher.md

@@ -25,7 +25,7 @@ You are a research specialist focused on **investigative rigor and source verifi
 
 **Always start with Perplexity Search.** Use the grounded search tool for your first sub-question:
 ```bash
-bun ~/.pal/skills/research/tools/perplexity-search.ts -- "<query>"
+pal cli skill run deep-research perplexity-search -- "<query>"
 ```
 
 - If it returns results → **continue using Perplexity Search** for remaining queries

package/assets/skills/analyze-pdf/SKILL.md

@@ -21,7 +21,7 @@ When the user asks to analyze, read, or extract information from a PDF:
 
 - **URL**: Use the `pdf-download` CLI tool to download and archive the PDF:
   ```bash
-  bun ~/.pal/skills/analyze-pdf/tools/pdf-download.ts -- <url> [--filename <name.pdf>]
+  pal cli skill run analyze-pdf pdf-download -- <url> [--filename <name.pdf>]
   ```
   The tool downloads the file, saves it to `memory/downloads/{YYYY}/{MM}/{DD}/{filename}.pdf`, and returns JSON with the saved `path`.
 
@@ -38,7 +38,7 @@ Do NOT install PDF processing tools (poppler, pdftotext, etc.) unless the user e
 If the Read tool fails to open the PDF (e.g. error mentioning `pdftoppm`, missing renderer, or unsupported format), fall back to the text-extraction CLI:
 
 ```bash
-bun ~/.pal/skills/analyze-pdf/tools/pdf-read.ts -- <path>
+pal cli skill run analyze-pdf pdf-read -- <path>
 ```
 
 Use the stdout output as the document content and proceed with the user's request as normal.

package/assets/skills/analyze-youtube/SKILL.md

@@ -23,7 +23,7 @@ When the user asks to analyze, summarize, or extract information from a YouTube
 Use the `youtube-analyze` CLI tool. It sends the video to Gemini, which processes both visual and audio content natively.
 
 ```bash
-bun ~/.pal/skills/analyze-youtube/tools/youtube-analyze.ts -- <youtube-url> [--prompt "your question"]
+pal cli skill run analyze-youtube youtube-analyze -- <youtube-url> [--prompt "your question"]
 ```
 
 - Without `--prompt`, it returns a structured summary with key insights, topics, people, and quotes.

package/assets/skills/consulting-report/SKILL.md

@@ -61,7 +61,7 @@ The component renders whatever `tunableLabel` you provide; the template never ha
 ### 1. Scaffold a new report
 
 ```bash
-bun ~/.pal/skills/consulting-report/tools/scaffold.ts <target-dir> \
+pal cli skill run consulting-report scaffold <target-dir> \
   [--client "Client Name"] [--title "Report Title"] [--no-install]
 ```
 
@@ -79,7 +79,7 @@ Static images go in `public/`; reference them from JSX as `<img src="/your-image
 ### 3. Live preview while authoring
 
 ```bash
-bun ~/.pal/skills/consulting-report/tools/dev.ts <report-dir>
+pal cli skill run consulting-report dev <report-dir>
 ```
 
 Wraps `bun run dev` in the report directory. Open the URL printed by Next, edit `app/page.tsx` or `lib/report-data.ts`, browser hot-reloads.
@@ -87,7 +87,7 @@ Wraps `bun run dev` in the report directory. Open the URL printed by Next, edit
 ### 4. Render the PDF
 
 ```bash
-node ~/.pal/skills/consulting-report/tools/generate-pdf.mjs <report-dir>
+pal cli skill run consulting-report generate-pdf.mjs <report-dir>
 ```
 
 Runs `next build` (which produces a static export at `out/`), then Playwright loads it via a tiny in-process HTTP server and prints the PDF with page-numbered header/footer. Output:
@@ -149,14 +149,14 @@ Run with **Node**, not Bun — Playwright's `chromium.launch()` hangs under Bun
 ## Demo
 
 ```bash
-node ~/.pal/skills/consulting-report/tools/generate-pdf.mjs ~/.pal/skills/consulting-report/demo
+pal cli skill run consulting-report generate-pdf.mjs ~/.pal/skills/consulting-report/demo
 ```
 
 Renders the bundled Acme Industries example end-to-end. Inspect the resulting PDF to see the full layout before authoring your own.
 
 ## Important
 
-- Run on Node (Playwright); the tool ships as a compiled `.mjs` so no `--experimental-strip-types` is needed
+- Runs on Node (Playwright); the tool ships as a compiled `.mjs`, and `pal cli skill run` picks Node for that extension so no `--experimental-strip-types` is needed
 - Bundled fonts come from Google Fonts via `next/font/google` — no licensing surface, no CDN at runtime, glyphs embedded at build time
 - Reports are disposable artifacts of `lib/report-data.ts` + `app/page.tsx`; commit the source, not the PDF
 - The scaffolder runs `bun install` inside the target by default — pass `--no-install` to skip

package/assets/skills/consulting-report/tools/dev.ts

@@ -3,7 +3,7 @@
 // consulting-report skill tool: launch the Next.js dev server for live preview.
 //
 // Usage:
-//   bun ~/.pal/skills/consulting-report/tools/dev.ts <report-dir>
+//   pal cli skill run consulting-report dev <report-dir>
 
 import { spawnSync } from "node:child_process";
 import { constants as fsConstants } from "node:fs";

package/assets/skills/consulting-report/tools/generate-pdf.ts

@@ -12,7 +12,7 @@
 // and is invoked as that .mjs: a .ts under node_modules can't be type-stripped by Node.
 //
 // Usage:
-//   node ~/.pal/skills/consulting-report/tools/generate-pdf.mjs <report-dir> [--pdf <out>] [--html <out>] [--skip-build]
+//   pal cli skill run consulting-report generate-pdf.mjs <report-dir> [--pdf <out>] [--html <out>] [--skip-build]
 
 import { spawnSync } from "node:child_process";
 import { createReadStream, constants as fsConstants, realpathSync } from "node:fs";

package/assets/skills/consulting-report/tools/scaffold.ts

@@ -4,7 +4,7 @@
 // and install Next.js dependencies.
 //
 // Usage:
-//   bun ~/.pal/skills/consulting-report/tools/scaffold.ts <target-dir> \
+//   pal cli skill run consulting-report scaffold <target-dir> \
 //        [--client "Client Name"] [--title "Report Title"] [--no-install]
 
 import { spawnSync } from "node:child_process";
@@ -100,7 +100,7 @@ async function run(argv: string[] = process.argv.slice(2)): Promise<void> {
   console.log(`  2. Edit lib/report-data.ts (metadata) and app/page.tsx (layout)`);
   console.log(`  3. Live preview:    bun run dev`);
   console.log(
-    `  4. Generate PDF:    node ~/.pal/skills/consulting-report/tools/generate-pdf.mjs ${opts.targetDir}`
+    `  4. Generate PDF:    pal cli skill run consulting-report generate-pdf.mjs ${opts.targetDir}`
   );
 }
 

package/assets/skills/create-pdf/SKILL.md

@@ -68,13 +68,13 @@ Invoke the skill tool. Flags:
 Single-file example:
 
 ```bash
-node ~/.pal/skills/create-pdf/tools/md-to-html-pdf.mjs /path/to/report.md --pdf /path/to/report.pdf
+pal cli skill run create-pdf md-to-html-pdf.mjs /path/to/report.md --pdf /path/to/report.pdf
 ```
 
 Multi-file example (after Step 2):
 
 ```bash
-node ~/.pal/skills/create-pdf/tools/md-to-html-pdf.mjs /tmp/combined.md --pdf /path/to/report.pdf --html /path/to/report.html
+pal cli skill run create-pdf md-to-html-pdf.mjs /tmp/combined.md --pdf /path/to/report.pdf --html /path/to/report.html
 ```
 
 The tool writes the self-contained HTML (inline CSS, UTF-8) and the PDF, and prints both paths + sizes on stdout.
@@ -95,7 +95,7 @@ Default styling (A4, 25mm margins, GitHub-ish look, table-friendly, page-break-a
 - `--header <html|file>` / `--footer <html|file>` — running header/footer on every page. The value is either an inline HTML string or a path to an HTML file. Templates may use Playwright's injected classes: `pageNumber`, `totalPages`, `date`, `title`, `url`.
 
 ```bash
-node ~/.pal/skills/create-pdf/tools/md-to-html-pdf.mjs report.md --pdf report.pdf \
+pal cli skill run create-pdf md-to-html-pdf.mjs report.md --pdf report.pdf \
   --margin 18mm \
   --header '<div style="font-size:9px;width:100%;text-align:center;color:#888">CONFIDENTIAL</div>' \
   --footer '<div style="font-size:9px;width:100%;text-align:right;padding-right:12mm;color:#888"><span class="pageNumber"></span>/<span class="totalPages"></span></div>'

package/assets/skills/create-pdf/tools/md-to-html-pdf.ts

@@ -10,7 +10,7 @@
 // and is invoked as that .mjs: a .ts under node_modules can't be type-stripped by Node.
 //
 // Usage:
-//   node ~/.pal/skills/create-pdf/tools/md-to-html-pdf.mjs <input.md> \
+//   pal cli skill run create-pdf md-to-html-pdf.mjs <input.md> \
 //     [--html <out.html>] [--pdf <out.pdf>] [--margin <css>] [--header <html|file>] [--footer <html|file>]
 //   --margin defaults to 25mm (all sides). --header/--footer accept inline HTML or a file path.
 

package/assets/skills/fyzz-chat-api/SKILL.md

@@ -21,19 +21,19 @@ When you need to access the user's Fyzz Chat conversations or projects, use the
 ### List conversations
 
 ```bash
-bun ~/.pal/skills/fyzz-chat-api/tools/fyzz-api.ts -- conversations [--limit 20] [--search "query"] [--project-id <id>] [--cursor <cursor>]
+pal cli skill run fyzz-chat-api fyzz-api -- conversations [--limit 20] [--search "query"] [--project-id <id>] [--cursor <cursor>]
 ```
 
 ### Get a single conversation with messages
 
 ```bash
-bun ~/.pal/skills/fyzz-chat-api/tools/fyzz-api.ts -- conversations <conversation-id>
+pal cli skill run fyzz-chat-api fyzz-api -- conversations <conversation-id>
 ```
 
 ### List projects
 
 ```bash
-bun ~/.pal/skills/fyzz-chat-api/tools/fyzz-api.ts -- projects
+pal cli skill run fyzz-chat-api fyzz-api -- projects
 ```
 
 ## Setup

package/assets/skills/onboarding/SKILL.md

@@ -57,7 +57,7 @@ Read a file only when you need the format of an existing entry, or when the user
 6. Write it with the telos tool, never by editing the file:
 
    ```bash
-   bun ~/.pal/skills/telos/tools/update-telos.ts <FILE> "<content>" "<description>"
+   pal cli skill run telos update-telos <FILE> "<content>" "<description>"
    ```
 
    `<FILE>` is the bare filename, `<description>` is a short line naming what the entry covers and the date it was given.

package/assets/skills/pal-analyze/SKILL.md

@@ -61,5 +61,5 @@ After showing results, ask:
 
 If yes to crystallizing:
 ```bash
-bun ~/.pal/tools/wisdom-frame.ts --domain <domain> --observation "<principle>"
+pal cli wisdom-frame --domain <domain> --observation "<principle>"
 ```

package/assets/skills/pal-reflect/SKILL.md

@@ -55,8 +55,8 @@ After showing results, say:
 If the user confirms or corrects something, use the opinion tool:
 ```bash
 # Confirmed
-bun ~/.pal/skills/opinion/tools/opinion.ts evidence "keywords" --confirmation "what they confirmed"
+pal cli skill run opinion opinion evidence "keywords" --confirmation "what they confirmed"
 
 # Corrected
-bun ~/.pal/skills/opinion/tools/opinion.ts evidence "keywords" --contradiction "what they corrected"
+pal cli skill run opinion opinion evidence "keywords" --contradiction "what they corrected"
 ```

package/assets/skills/playwright/SKILL.md

@@ -36,7 +36,7 @@ If neither engine is usable, the tool prints `NO_PLAYWRIGHT_CLI` and exits non-z
 2. Run the tool (it prints the absolute PNG path as its last stdout line):
 
    ```bash
-   bun ~/.pal/skills/playwright/tools/shot.ts <url> \
+   pal cli skill run playwright shot <url> \
      [--viewport 1440x900] [--full-page] [--selector "<css>"] [-o <out.png>]
    ```
 

package/assets/skills/playwright/tools/shot.ts

@@ -9,7 +9,7 @@
 // caller (SKILL.md) can fall back to the Playwright MCP.
 //
 // Runs under Bun on every OS:
-//   bun ~/.pal/skills/playwright/tools/shot.ts <url> [opts]
+//   pal cli skill run playwright shot <url> [opts]
 // It used to ship a Node-compiled .mjs sibling because chromium.launch hung under Bun on
 // Windows; bun 1.4.0 fixes that, so the Node hop and the build step are gone.
 

package/assets/skills/presentation/README.md

@@ -6,12 +6,12 @@ See `SKILL.md` for the full workflow. TL;DR:
 
 ```bash
 # one-time per brand
-bun ~/.pal/skills/presentation/tools/setup-template.ts
+pal cli skill run presentation setup-template
 
 # per deck
-bun ~/.pal/skills/presentation/tools/new-deck.ts ~/decks/my-talk --template my-brand
+pal cli skill run presentation new-deck ~/decks/my-talk --template my-brand
 $EDITOR ~/decks/my-talk/slides/
-bun ~/.pal/skills/presentation/tools/build.ts ~/decks/my-talk
+pal cli skill run presentation build ~/decks/my-talk
 # → ./my-talk/my-talk.html (and ./my-talk/my-talk.md — concatenated source)
 # open the .html in your browser; refresh after each rebuild
 ```

package/assets/skills/presentation/SKILL.md

@@ -32,14 +32,14 @@ Renders a deck folder (markdown + config) to a single self-contained HTML presen
 
 Interactive:
 ```bash
-bun ~/.pal/skills/presentation/tools/setup-template.ts
+pal cli skill run presentation setup-template
 ```
 
 Walks through 9 fields: name, storage path, logo file, primary color, accent color, footer text, logo placement, fonts, aspect ratio. Optional: generate a showcase deck demonstrating every layout.
 
 Non-interactive (Claude can drive it):
 ```bash
-bun ~/.pal/skills/presentation/tools/setup-template.ts \
+pal cli skill run presentation setup-template \
   --name <slug> \
   --logo <abs-path-to-logo.svg> \
   --primary "#0E1335" \
@@ -57,7 +57,7 @@ Defaults if omitted: accent = derived complementary of primary; logo-placement =
 ### Step 1: Scaffold a deck
 
 ```bash
-bun ~/.pal/skills/presentation/tools/new-deck.ts <deck-dir> --template <name> [--title "Deck title"]
+pal cli skill run presentation new-deck <deck-dir> --template <name> [--title "Deck title"]
 ```
 
 If `--template` is omitted and only one template is registered, that one is used. If multiple are registered, the command lists them and exits. Adds `--showcase` to scaffold a demo deck with every layout exercised.
@@ -79,7 +79,7 @@ Backwards compatible: if `slides/` doesn't exist, the build falls back to a sing
 ### Step 3: Build
 
 ```bash
-bun ~/.pal/skills/presentation/tools/build.ts <deck-dir> [--out <dir>] [--force]
+pal cli skill run presentation build <deck-dir> [--out <dir>] [--force]
 ```
 
 Output files (where `<deck-name>` = basename of `<deck-dir>`):
@@ -96,7 +96,7 @@ Override: pass `--out <dir>` to redirect elsewhere. When `--out` is *not* the de
 ### Step 3.5 (optional but recommended): Lint with the doctor
 
 ```bash
-bun ~/.pal/skills/presentation/tools/doctor.ts <deck-dir> [--strict]
+pal cli skill run presentation doctor <deck-dir> [--strict]
 ```
 
 Catches authoring failures before you ever open the browser:
@@ -346,13 +346,13 @@ Print this block as the closing of any turn that touches slides:
 ```bash
 # lint — catches overflow, missing assets, layout-content mismatches
 # bash / PowerShell / Git Bash:
-bun ~/.pal/skills/presentation/tools/doctor.ts <deck-dir>
+pal cli skill run presentation doctor <deck-dir>
 # Windows cmd.exe:
 bun %USERPROFILE%\.pal\skills\presentation\tools\doctor.ts <deck-dir>
 
 # build — writes <cwd>/<deck-name>/<deck-name>.{html,md}
 # bash / PowerShell / Git Bash:
-bun ~/.pal/skills/presentation/tools/build.ts <deck-dir>
+pal cli skill run presentation build <deck-dir>
 # Windows cmd.exe (no ~ expansion):
 bun %USERPROFILE%\.pal\skills\presentation\tools\build.ts <deck-dir>
 
@@ -365,7 +365,7 @@ Substitute `<deck-dir>` with the actual deck path. Do this even when you also ra
 ## Other commands
 
 ```bash
-bun ~/.pal/skills/presentation/tools/list-templates.ts
+pal cli skill run presentation list-templates
 ```
 
 Prints all registered templates with their primary color and storage path.

package/assets/skills/presentation/template/README.md

@@ -2,7 +2,7 @@
 
 One markdown file per slide under `slides/`. Files are concatenated at build time in filename order (use leading zeros: `001.md`, `002.md`, …). **Don't put `---` separators inside slide files** — the separator is added between files at build time.
 
-Build: `bun ~/.pal/skills/presentation/tools/build.ts .`
+Build: `pal cli skill run presentation build .`
 
 Output lands at `<cwd>/<deck-name>/<deck-name>.html` — open it in your browser and refresh after each rebuild.
 

package/assets/skills/presentation/tools/new-deck.ts

@@ -103,7 +103,7 @@ lang: en
   console.log(`  slides:     ${sourceFiles.length} file(s) in slides/`);
   console.log(`\nNext:`);
   console.log(`  $EDITOR ${slidesDir}/`);
-  console.log(`  bun ~/.pal/skills/presentation/tools/build.ts ${target}`);
+  console.log(`  pal cli skill run presentation build ${target}`);
   console.log(`  # output → ${target}/${slug}.html  (override with --out <dir>)`);
 }
 

package/assets/skills/presentation/tools/setup-template.ts

@@ -344,7 +344,7 @@ aspect: "${aspect}"
   }
 
   console.log(
-    `\nNext: bun ~/.pal/skills/presentation/tools/new-deck.ts <deck-dir> --template ${name}`
+    `\nNext: pal cli skill run presentation new-deck <deck-dir> --template ${name}`
   );
 }
 

package/assets/skills/projects/SKILL.md

@@ -30,7 +30,7 @@ Manage the user's project registry. Each project lives at `~/.pal/memory/project
 All operations go through the canonical CLI:
 
 ```bash
-bun ~/.pal/tools/project.ts <command> [args]
+pal cli project <command> [args]
 ```
 
 Output is JSON.
@@ -40,7 +40,7 @@ Output is JSON.
 | `list` | All registered projects with status, path, updated, stale flag, and counts |
 | `create [name] [--path PATH] [--objectives "a;b;c"] [--serves KIND] [--serves-note "..."]` | Register a project. Defaults: name=basename(cwd), path=cwd. Slug must be `[a-z0-9_-]+`. KIND is `goal`, `revenue` or `fun` |
 | `serves <name> <goal\|revenue\|fun> [note]` | Set what the project is for, after the fact |
-| `resume <name>` | Print the full project ISA — all frontmatter and body sections |
+| `resume <name>` | Print the lean project view — frontmatter, body sections, and open-ISC titles (full ISC text via `show-isc`) |
 | `add-next <name> "text"` | Append a next step (array, instantly appendable) |
 | `add-blocker <name> "text"` | Append a blocker (array, instantly appendable) |
 | `add-decision <name> "decision" "rationale"` | Log a timestamped decision entry to the Decisions section |
@@ -48,6 +48,15 @@ Output is JSON.
 | `rm-next \| rm-blocker <name> <index>` | Remove a next/blocker entry by zero-based index |
 | `update-section <name> <section> "content"` | Set an ISA body section (problem, goal, criteria, vision, constraints, out_of_scope, context, decisions, changelog) |
 | `criteria <name>` | Print the Criteria section (verifiable success conditions) |
+| `set-path <name> <new-path>` | Update the registered path |
+| `add-isc <name> "title"` | Append a new open ISC to Criteria |
+| `list-isc <name> [--all \| --closed \| --retired]` | List open ISCs (default), or the archived/retired sets |
+| `show-isc <name> <id>` | Print one ISC's full text |
+| `edit-isc <name> <id> "new text"` | Rewrite an ISC's wording, keeping its id and state |
+| `complete-isc \| reopen-isc <name> <id>` | Archive a finished ISC, or pull one back into the open set |
+| `retire-isc <name> <id> [--by <id>]` | Close an ISC that stopped being valid, not as done |
+| `prune-isc <name>` | Sweep legacy done ISCs from Criteria into the Changelog archive |
+| `scaffold-task-isa <title>` / `complete-task-isa <slug>` | Create and close a one-shot task ISA in `memory/work/` |
 | `isa-init <name>` | Mark a project as ISA-initialized |
 | `complete <name>` / `archive <name>` / `pause <name>` / `unpause <name>` | Status transitions |
 | `rm <name>` | Delete the project directory entirely |
@@ -115,7 +124,7 @@ When the user describes next steps, blockers, or decisions during normal work, i
 User: "store under <project> that a reference implementation exists in this repo"
 → Identify the project from `list` (or by name)
 → Durable reference, not a task → update-section
-→ bun ~/.pal/tools/project.ts update-section <slug> context "Reference implementation lives in this repo"
+→ pal cli project update-section <slug> context "Reference implementation lives in this repo"
 ```
 
 **Registering the current repo**
@@ -123,32 +132,32 @@ User: "store under <project> that a reference implementation exists in this repo
 User: "track this project"
 → Default name from cwd basename, confirm with user
 → Ask what it serves: a goal, a way it could pay, or fun
-→ bun ~/.pal/tools/project.ts create --path "$(pwd)" --objectives "first objective; second objective" --serves revenue --serves-note "could be sold as a service"
+→ pal cli project create --path "$(pwd)" --objectives "first objective; second objective" --serves revenue --serves-note "could be sold as a service"
 ```
 
 **Correcting what a project is for**
 ```
 User: "<project> isn't a toy, it's the thing I'd actually sell"
-→ bun ~/.pal/tools/project.ts serves <slug> revenue "the one I would sell"
+→ pal cli project serves <slug> revenue "the one I would sell"
 ```
 
 **Logging a decision**
 ```
 User: "we decided <decision> because <reason>"
-→ bun ~/.pal/tools/project.ts add-decision <slug> "<decision>" "<reason>"
+→ pal cli project add-decision <slug> "<decision>" "<reason>"
 ```
 
 **Setting the goal and criteria**
 ```
 User: "set the goal for pal to 'ship ISA support with full test coverage'"
-→ bun ~/.pal/tools/project.ts update-section pal goal "ship ISA support with full test coverage"
+→ pal cli project update-section pal goal "ship ISA support with full test coverage"
 ```
 
 **Completing a project**
 ```
 User: "mark <project> as complete"
 → Confirm
-→ bun ~/.pal/tools/project.ts complete <slug>
+→ pal cli project complete <slug>
 ```
 
 ## Anti-patterns

package/assets/skills/telos/SKILL.md

@@ -43,7 +43,7 @@ Read the file directly from `~/.pal/telos/` when the user asks about any area. S
 Appends content, creates backup, logs the change:
 
 ```bash
-bun ~/.pal/skills/telos/tools/update-telos.ts <FILE> "<content>" "<description>"
+pal cli skill run telos update-telos <FILE> "<content>" "<description>"
 ```
 
 ## Routing

package/assets/templates/AGENTS.md.template

@@ -55,7 +55,7 @@ Start your response with the following header in this mode:
 
 - **Mandatory output format** — Every response MUST use exactly one of the output formats above (ALGORITHM, NATIVE, or MINIMAL). No freeform output.
 - **Response format before questions** — Always complete the current response format output FIRST, then invoke AskUserQuestion at the end.
-- **Named project → resume first** — Whenever the user references a project by name in ANY mode, run `bun ~/.pal/tools/project.ts resume <slug>` before any file search or task execution. The project record has paths, context, and handoff notes that make manual searching unnecessary.
+- **Named project → resume first** — Whenever the user references a project by name in ANY mode, run `pal cli project resume <slug>` before any file search or task execution. The project record has paths, context, and handoff notes that make manual searching unnecessary.
 
 ---
 

package/assets/templates/PAL/ALGORITHM.md

@@ -41,10 +41,10 @@ Thinking-only. No tool calls except context recovery (Grep/Glob/Read).
 ```bash
 # If cwd matches a registered project — read its open ISCs (Ideal State Criteria):
 # list-isc returns OPEN criteria by default; pass --all or --closed to see finished ones.
-bun ~/.pal/tools/project.ts list-isc <project-name>
+pal cli project list-isc <project-name>
 
 # If this is ad-hoc work with no registered project — scaffold a task ISA:
-bun ~/.pal/tools/project.ts scaffold-task-isa "<task title>"
+pal cli project scaffold-task-isa "<task title>"
 ```
 
 Surface any open ISCs as live context: they are unfinished criteria from prior sessions. New criteria defined in this session extend them (use `add-isc`), not replace them.
@@ -243,7 +243,7 @@ Focus: reasoning approach, problem decomposition, anticipation, blind spots.
 **2. Reflection Log** — record algorithm performance:
 
 ```bash
-bun ~/.pal/tools/algorithm-reflect.ts --task "description" --criteria N --passed N --failed N --sentiment 1-10 \
+pal cli algorithm-reflect --task "description" --criteria N --passed N --failed N --sentiment 1-10 \
   --q1 "self reflection" --q2 "algorithm reflection" --q3 "AI reflection" --scope general
 ```
 
@@ -255,16 +255,16 @@ The goal is behavioral intelligence about the user, not a session log. Session l
 
 ```bash
 # Opinion — behavioral observation about the user (what you noticed about how they work):
-bun ~/.pal/tools/relationship-note.ts --o "User prefers reviewing existing code before adding anything new" --confidence 0.80
+pal cli relationship-note --o "User prefers reviewing existing code before adding anything new" --confidence 0.80
 
 # World fact — objective fact about the user's situation (tech stack, project state, context):
-bun ~/.pal/tools/relationship-note.ts --w "User is building a backend service in TypeScript with Bun"
+pal cli relationship-note --w "User is building a backend service in TypeScript with Bun"
 
 # Session diary — what the agent did this session (first-person, specific):
-bun ~/.pal/tools/relationship-note.ts --b "Refactored the auth middleware to support refresh token rotation"
+pal cli relationship-note --b "Refactored the auth middleware to support refresh token rotation"
 
 # Multiple notes in one call:
-bun ~/.pal/tools/relationship-note.ts --o "User prefers one verified change at a time, not batches" --confidence 0.80 --b "Fixed the path-normalization bug in the hook merge logic"
+pal cli relationship-note --o "User prefers one verified change at a time, not batches" --confidence 0.80 --b "Fixed the path-normalization bug in the hook merge logic"
 ```
 
 **O (Opinion) — what to write:**
@@ -291,10 +291,10 @@ Skip only if the session was a trivial lookup or typo fix (same rule as step 2).
 
 ```bash
 # Work still in progress:
-bun ~/.pal/tools/handoff-note.ts --title "what we were doing" --text "what remains, decisions made, next steps"
+pal cli handoff-note --title "what we were doing" --text "what remains, decisions made, next steps"
 
 # Work finished — clear any previous in-progress handoff:
-bun ~/.pal/tools/handoff-note.ts --done --title "what we completed"
+pal cli handoff-note --done --title "what we completed"
 ```
 
 - Write if anything is left mid-flight: unfinished implementation, open decision, partially debugged issue
@@ -306,21 +306,21 @@ bun ~/.pal/tools/handoff-note.ts --done --title "what we completed"
 **Project work** — use ISCs, not threads:
 ```bash
 # Close completed ISCs:
-bun ~/.pal/tools/project.ts complete-isc <project-name> <id>
+pal cli project complete-isc <project-name> <id>
 
 # Open new ISCs for unfinished work:
-bun ~/.pal/tools/project.ts add-isc <project-name> "what remains"
+pal cli project add-isc <project-name> "what remains"
 ```
 
 **Task ISA (one-shot work)** — mark complete when done:
 ```bash
-bun ~/.pal/tools/project.ts complete-task-isa <slug>
+pal cli project complete-task-isa <slug>
 ```
 
 **Cross-project or non-project follow-ups** — use threads:
 ```bash
-bun ~/.pal/tools/thread.ts --add --title "brief title" --context "why it matters, what needs to happen"
-bun ~/.pal/tools/thread.ts --resolve --id <id>
+pal cli thread --add --title "brief title" --context "why it matters, what needs to happen"
+pal cli thread --resolve --id <id>
 ```
 
 **6. Opinion capture** — scan the conversation for moments where the user:
@@ -331,13 +331,13 @@ bun ~/.pal/tools/thread.ts --resolve --id <id>
 For each, invoke the opinion tool:
 ```bash
 # User confirmed a preference
-bun ~/.pal/skills/opinion/tools/opinion.ts evidence "matching keywords" --confirmation "what they confirmed"
+pal cli skill run opinion opinion evidence "matching keywords" --confirmation "what they confirmed"
 
 # User corrected a preference
-bun ~/.pal/skills/opinion/tools/opinion.ts evidence "matching keywords" --contradiction "what they corrected"
+pal cli skill run opinion opinion evidence "matching keywords" --contradiction "what they corrected"
 
 # New pattern observed (no existing opinion matches)
-bun ~/.pal/skills/opinion/tools/opinion.ts add "the preference" --category communication|technical|workflow|general
+pal cli skill run opinion opinion add "the preference" --category communication|technical|workflow|general
 ```
 
 Skip if nothing in the conversation touched preferences or working style.
@@ -345,7 +345,7 @@ Skip if nothing in the conversation touched preferences or working style.
 **7. Wisdom Frame** (Extended+ only) — if the session produced a genuine, reusable insight:
 
 ```bash
-bun ~/.pal/tools/wisdom-frame.ts --domain <domain> --observation "insight" [--type principle|contextual-rule|anti-pattern|evolution]
+pal cli wisdom-frame --domain <domain> --observation "insight" [--type principle|contextual-rule|anti-pattern|evolution]
 ```
 
 Domains: `development`, `workflow`, `communication`, `infrastructure`, `integration`, or any fitting domain.

package/assets/templates/PAL/PROJECT_LIFECYCLE.md

@@ -1,6 +1,6 @@
 # Project Lifecycle
 
-You (the AI) own the project lifecycle. Project state lives in `~/.pal/memory/state/progress/{slug}.json`, one file per project, managed via `bun ~/.pal/tools/project.ts`. Active projects are auto-injected into every SessionStart context regardless of cwd.
+You (the AI) own the project lifecycle. Project state lives in `~/.pal/memory/state/progress/{slug}.json`, one file per project, managed via `pal cli project`. Active projects are auto-injected into every SessionStart context regardless of cwd.
 
 ## When to invoke the CLI
 

package/assets/templates/PAL/WORK_TRACKING.md

@@ -4,4 +4,4 @@ PAL tracks your work across sessions in `memory/state/sessions.json` (auto-captu
 
 ## Projects
 
-Projects are managed via the `/projects` skill. State lives in `~/.pal/memory/state/progress/{slug}.json`, one file per project. Inspect with `bun ~/.pal/tools/project.ts list`; manage via `~/.pal/docs/PROJECT_LIFECYCLE.md`.
+Projects are managed via the `/projects` skill. State lives in `~/.pal/memory/state/progress/{slug}.json`, one file per project. Inspect with `pal cli project list`; manage via `~/.pal/docs/PROJECT_LIFECYCLE.md`.