atris 3.74.4 → 3.74.5

package/atris/skills/autopilot/SKILL.md

@@ -10,7 +10,7 @@ tags:
   - flow
 ---
 
-> **The two-engine architecture:** /flow chains forward from identity (who you are → next move). /endgame chains backward from horizon (where you're going → next move). They meet at the same intersection — the next thing /autopilot should ship. Each tick reads both sides and shows them in the visual status block.
+> **The two-engine architecture:** /flow chains forward from identity (who you are → next move). /endgame chains backward from horizon (where you're going → next move). They meet at the same intersection: the next thing /autopilot should ship. Each tick reads both sides and shows them in the visual status block.
 
 ## Visual status block
 
@@ -34,7 +34,7 @@ If identity is missing, edit PERSONA.md. If no horizon is active, /endgame seeds
 
 # /autopilot
 
-Runs ONE plan→do→review tick anchored to the current endgame. If no endgame is active, seeds one from the inbox or wiki signals first. Not a recurring loop — call `/loop` for that.
+Runs ONE plan→do→review tick anchored to the current endgame. If no endgame is active, seeds one from the inbox or wiki signals first. Not a recurring loop. Call `/loop` for that.
 
 ## When to use
 
@@ -43,10 +43,10 @@ Runs ONE plan→do→review tick anchored to the current endgame. If no endgame
 
 ## What ONE tick does
 
-1. **Read state** — load `atris/TODO.md`. Look for the `## Endgame` section header and `[endgame]`-tagged tasks in `## Backlog`.
-2. **Boundary check** — if no `## Endgame` section exists, OR every `[endgame]`-tagged task in Backlog is done/missing, invoke `/endgame` first to seed the next horizon. `/endgame` will write a new `## Endgame` section + tagged backlog tasks to TODO.md.
-3. **Execute** — run `atris autopilot --auto --iterations=1` via Bash. The CLI prefers `[endgame]`-tagged backlog tasks (priority 0) over reactive signals (priority 1+). One task per tick.
-4. **Stop** — show the output, do not start a conversation, do not chain. The next tick is the cron's job.
+1. **Read state**: load `atris/TODO.md`. Look for the `## Endgame` section header and `[endgame]`-tagged tasks in `## Backlog`.
+2. **Boundary check**: if no `## Endgame` section exists, OR every `[endgame]`-tagged task in Backlog is done/missing, invoke `/endgame` first to seed the next horizon. `/endgame` will write a new `## Endgame` section + tagged backlog tasks to TODO.md.
+3. **Execute**: run `atris autopilot --auto --iterations=1` via Bash. The CLI prefers `[endgame]`-tagged backlog tasks (priority 0) over reactive signals (priority 1+). One task per tick.
+4. **Stop**: show the output, do not start a conversation, do not chain. The next tick is the cron's job.
 
 ## How to invoke
 
@@ -76,9 +76,9 @@ no endgame anywhere, no inbox, clean   → loop journals "nothing left", stops
 
 ## Variants
 
-- `atris autopilot --dry-run` — preview what it would do, do not execute
-- `atris autopilot --auto --iterations=N` — run up to N ticks back-to-back (still one task per tick, boundary checks between)
-- `atris autopilot "<task description>"` — seed a new inbox item, then run
+- `atris autopilot --dry-run`: preview what it would do, do not execute
+- `atris autopilot --auto --iterations=N`: run up to N ticks back-to-back (still one task per tick, boundary checks between)
+- `atris autopilot "<task description>"`: seed a new inbox item, then run
 
 ## Autonomous mode
 
@@ -95,6 +95,6 @@ If the user wants this to fire on a recurring schedule, invoke `/loop` instead.
 - Always show *why* before executing.
 - Stop after the first tick. Do not chain. Chaining is `/loop`'s job.
 - Endgame tasks always preferred over reactive signals.
-- At every boundary (no current endgame OR all done), reassess via `/endgame` — read inbox/wiki/logs, pick the next horizon, do not just run forever.
+- At every boundary (no current endgame OR all done), reassess via `/endgame`: read inbox/wiki/logs, pick the next horizon, do not just run forever.
 - If a tick fails, halt and journal the failure. Do not pretend it worked.
 - The CLI writes the heartbeat Notes block. Do not hand-write tick summaries to the journal.

package/atris/skills/blocks/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: blocks
-description: Author Atris block documents — the live markdown docs the Atris (obelisk) editor renders, with charts, gauges, metrics, tables, and images. One doc is a beautiful one-pager; add slide breaks and it presents as a deck. Use when creating or editing an Atris doc, deck, one-pager, report, or anything opened in the Atris markdown editor.
+description: Author Atris block documents, the live markdown docs the Atris (obelisk) editor renders, with charts, gauges, metrics, tables, and images. One doc is a beautiful one-pager; add slide breaks and it presents as a deck. Use when creating or editing an Atris doc, deck, one-pager, report, or anything opened in the Atris markdown editor.
 version: 1.0.0
 tags:
   - blocks
@@ -14,7 +14,7 @@ tags:
 
 Author documents the Atris desktop editor renders as **live blocks**: charts, gauges,
 metric cards, tables, and images inside plain markdown. Same file is a one-pager or a
-deck — the only difference is slide breaks.
+deck. The only difference is slide breaks.
 
 The whole thing is one `.md` file. Prose is prose. A fenced code block with a known
 language becomes a live visual. A `---` on its own line starts a new slide.
@@ -25,7 +25,7 @@ The renderer lives in the Atris desktop app (obelisk repo), at
 `project-obelisk/src/lib/markdownEditorIo.cjs` (engine) and `renderEmbedBlockHtml` in
 `project-obelisk/src/components/AppExperience.tsx` (block defs). Present mode:
 `MarkdownPresentView.tsx`. Working templates: `project-obelisk/atris/decks/`
-(`blocks-demo.md`, `generative-blocks.md`). Validate against the engine (below) —
+(`blocks-demo.md`, `generative-blocks.md`). Validate against the engine (below):
 it's the source of truth, no app needed.
 
 ## Doc shape
@@ -36,27 +36,27 @@ it's the source of truth, no app needed.
 
 ## The blocks
 
-Each is a fenced code block. The body must be **valid JSON** — invalid JSON silently
+Each is a fenced code block. The body must be **valid JSON**. Invalid JSON silently
 falls back to a plain code block (no error), so validate. Colors/accents are hex strings.
 
-### metric — a stat card
+### metric: a stat card
 ```metric
 {"label":"MRR from paying customers","value":"$20K+","accent":"#f0a13a"}
 ```
 `label`, `value` (string), optional `accent`. Put several in a row; they stack as cards.
 
-### gauge — a ring toward a target
+### gauge: a ring toward a target
 ```gauge
 {"title":"Ledger reconciled","value":100,"max":100,"caption":"ties back to the board, to the penny"}
 ```
 `title` (or `label`), `value`, `max`, optional `caption`, `accent`.
 
-### chart — line / bar / area / donut
+### chart: line / bar / area / donut
 Single series (friendly `data` form):
 ```chart
 {"kind":"bar","title":"Weekly active sessions","data":[{"label":"W1","value":12},{"label":"W2","value":19},{"label":"W3","value":27}]}
 ```
-Multiple series (line, with colors — great for "rent rises vs own falls"):
+Multiple series (line, with colors, great for "rent rises vs own falls"):
 ```chart
 {"kind":"line","title":"Cost per unit of work","series":[{"name":"Rent","color":"#d16a5a","points":[{"label":"Y1","value":40},{"label":"Y2","value":88}]},{"name":"Own","color":"#3f9668","points":[{"label":"Y1","value":38},{"label":"Y2","value":12}]}],"caption":"same work, opposite curves"}
 ```
@@ -66,7 +66,7 @@ Donut (first series' points become slices; center shows the total):
 ```
 `kind`: `line` | `bar` | `area` | `donut`. Points are `{"label","value"}`. Optional `unit`, `caption`. A chart with no points renders nothing, so always include data.
 
-### table — an editable grid
+### table: an editable grid
 Plain GFM. Renders as a clean grid, round-trips verbatim. First column can be a blank header for a comparison.
 ```
 | | Rent | Own |
@@ -75,13 +75,13 @@ Plain GFM. Renders as a clean grid, round-trips verbatim. First column can be a
 | Data | Their building | Your walls |
 ```
 
-### image — a picture
+### image: a picture
 ```image
 {"url":"https://...","alt":"..."}
 ```
-A data: or https: URL renders immediately. Without one, the editor shows a placeholder and generates (non-deterministic — don't rely on generation for something you're about to present).
+A data: or https: URL renders immediately. Without one, the editor shows a placeholder and generates (non-deterministic, so don't rely on generation for something you're about to present).
 
-### ask — an embedded AI answer
+### ask: an embedded AI answer
 ```ask
 {"output":"the answer, as markdown"}
 ```
@@ -92,12 +92,12 @@ Renders `output` as prose. Authored by the editor's in-app AI; rarely hand-writt
 - **No side-by-side columns by hand.** `md-cols` only parses as a single unbroken line and breaks the moment you add newlines/bullets. Use a **table** for comparisons. Columns are a UI-only construct.
 - **Invalid JSON = silent fallback** to a code block. No error is thrown. Validate.
 - **A chart needs data points** or it renders empty.
-- **Raw HTML is sanitized** — inline `style=` and `<style>`/`<script>` are stripped. You cannot brand a doc with custom CSS; it wears the editor's theme. Design with the blocks, not CSS.
+- **Raw HTML is sanitized**: inline `style=` and `<style>`/`<script>` are stripped. You cannot brand a doc with custom CSS; it wears the editor's theme. Design with the blocks, not CSS.
 - **`---` is a hard slide break.** A stray `---` inside a slide splits it. GFM table delimiters (`| --- |`) are safe.
 
 ## Validate before you trust it
 
-Run the doc through the real engine — no app needed. This auto-finds obelisk's engine
+Run the doc through the real engine, no app needed. This auto-finds obelisk's engine
 from any project in the workspace:
 
 ```bash
@@ -121,7 +121,7 @@ slides.forEach((s,i)=>{
 
 ## Open / present / export
 
-- **Open:** in the app, Cmd+P → type the filename → Enter. (Quick Open searches the open project only — the file must live under it.)
+- **Open:** in the app, Cmd+P → type the filename → Enter. (Quick Open searches the open project only, so the file must live under it.)
 - **Edit a block:** click the chart/number; a form opens to change values. No JSON editing needed.
 - **Insert a block:** type `/` for the block menu.
 - **Present:** the Present button (splits on `---`); arrows/space to move, Esc to exit.
@@ -131,4 +131,4 @@ slides.forEach((s,i)=>{
 
 Lead with the blocks that carry meaning, one hero visual per slide. A number that
 matters is a `metric`. A trend or a crossing is a `chart`. A comparison is a `table`.
-Everything else is tight prose. Don't decorate — the doc should read clean at a squint.
+Everything else is tight prose. Don't decorate. The doc should read clean at a squint.

package/atris/skills/calendar/SKILL.md

@@ -154,7 +154,7 @@ curl -s -X POST "https://api.atris.ai/api/integrations/google-calendar/events" \
   }'
 ```
 
-**IMPORTANT:** Use `POST` to create events. Do NOT use `PUT` — that is for updating existing events.
+**IMPORTANT:** Use `POST` to create events. Do NOT use `PUT`. That is for updating existing events.
 
 ### Update Event
 ```bash

package/atris/skills/clawhub/atris/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: atris
-description: "Codebase intelligence — generates structured navigation maps with file:line references so agents stop re-scanning the same files every session. Use when exploring code, answering 'where is X?', or onboarding to a new codebase."
+description: "Codebase intelligence: generates structured navigation maps with file:line references so agents stop re-scanning the same files every session. Use when exploring code, answering 'where is X?', or onboarding to a new codebase."
 version: 1.1.0
 requires:
   bins:
@@ -13,7 +13,7 @@ tags:
   - context-management
 ---
 
-# Atris — Codebase Intelligence
+# Atris: Codebase Intelligence
 
 Maintain a structured map of the codebase with exact file:line references. One scan, permanent knowledge. Saves 80-95% of tokens on code exploration.
 
@@ -63,7 +63,7 @@ rg "listen|createServer|app\.start|if __name__" --line-number
 ## MAP.md structure
 
 ```markdown
-# MAP.md — [Project Name] Navigation Guide
+# MAP.md: [Project Name] Navigation Guide
 
 > Generated by Atris | Last updated: YYYY-MM-DD
 
@@ -98,7 +98,7 @@ Flag high-impact files with why they matter and key functions with line numbers.
 
 ### Entry Points
 
-How execution flows — dev server startup, request lifecycle, build pipeline.
+How execution flows: dev server startup, request lifecycle, build pipeline.
 
 ## Keeping it fresh
 

package/atris/skills/clawhub/philosophy-of-work/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: philosophy-of-work
-description: "Install Atris — the open-source operating system for AI work. Gives agents MAP-first navigation, durable tasks, proof on disk, and a plan→do→review loop. Use when starting in a repo, onboarding, or when chat context keeps dying."
+description: "Install Atris: the open-source operating system for AI work. Gives agents MAP-first navigation, durable tasks, proof on disk, and a plan→do→review loop. Use when starting in a repo, onboarding, or when chat context keeps dying."
 version: 1.0.0
 requires:
   bins:
@@ -15,7 +15,7 @@ tags:
 
 # Philosophy of Work (Atris)
 
-Humans are building a **philosophy of work** — how companies and agents should run together.
+Humans are building a **philosophy of work**: how companies and agents should run together.
 
 The **bible is open source** (CLI + `atris.md`). The **school** is optional cloud + humans who help design the system.
 
@@ -33,10 +33,10 @@ Read `atris/atris.md` first. Follow it exactly.
 
 ## Why switch
 
-- **MAP.md** — stop re-scanning; file:line navigation
-- **atris task** — work survives sessions; claim → ship → proof
-- **Missions** — overnight loops with verifiers
-- **Honest handoffs** — next agent reads disk, not your chat
+- **MAP.md**: stop re-scanning; file:line navigation
+- **atris task**: work survives sessions; claim → ship → proof
+- **Missions**: overnight loops with verifiers
+- **Honest handoffs**: next agent reads disk, not your chat
 
 ## Agent contract (4 artifacts)
 
@@ -47,7 +47,7 @@ Read `atris/atris.md` first. Follow it exactly.
 | Change | small git diff |
 | Proof | `atris task ready <id> --proof "..."` |
 
-Do not `atris task accept` — humans only.
+Do not `atris task accept`. Humans only.
 
 ## Full letter
 

package/atris/skills/create-app/SKILL.md

@@ -104,7 +104,7 @@ unset secret_val
 echo "Saved locally."
 ```
 
-Register the key in the web UI (manifest only — no value sent):
+Register the key in the web UI (manifest only, no value sent):
 ```bash
 curl -s -X POST "https://api.atris.ai/api/apps/SLUG/secrets/KEY_NAME/register-local" \
   -H "Authorization: Bearer $TOKEN"
@@ -222,9 +222,9 @@ curl -s -X POST "https://api.atris.ai/api/scheduled-tasks" \
 ```
 
 Common schedules:
-- `"0 8 * * *"` — daily at 8am UTC
-- `"0 */6 * * *"` — every 6 hours
-- `"0 9 * * 1"` — weekly Monday 9am UTC
+- `"0 8 * * *"`: daily at 8am UTC
+- `"0 */6 * * *"`: every 6 hours
+- `"0 9 * * 1"`: weekly Monday 9am UTC
 
 **Skip this step** for on-demand apps (manual trigger only) or chat apps.
 
@@ -270,20 +270,20 @@ Tell the user:
 All endpoints use the app's `share_token` as `{slug}`.
 
 ```
-POST   /api/apps/{slug}/trigger          — run the app now
-POST   /api/apps/{slug}/ingest           — push data in
-POST   /api/apps/{slug}/ingest/batch     — push multiple items
-GET    /api/apps/{slug}/data             — read stored data
-GET    /api/apps/{slug}/data/{collection} — read specific collection
-GET    /api/apps/{slug}/status           — health, last run, next run
-GET    /api/apps/{slug}/runs             — execution history
-GET    /api/apps/{slug}/runs/{run_id}    — single run details
-PUT    /api/apps/{slug}/secrets/{key}    — store an API key (owner only)
-GET    /api/apps/{slug}/secrets          — list key names (owner only)
-DELETE /api/apps/{slug}/secrets/{key}    — remove a key (owner only)
-POST   /api/apps/{slug}/members          — add agent operator
-GET    /api/apps/{slug}/members          — list members
-DELETE /api/apps/{slug}/members/{agent_id} — remove member
+POST   /api/apps/{slug}/trigger          : run the app now
+POST   /api/apps/{slug}/ingest           : push data in
+POST   /api/apps/{slug}/ingest/batch     : push multiple items
+GET    /api/apps/{slug}/data             : read stored data
+GET    /api/apps/{slug}/data/{collection} : read specific collection
+GET    /api/apps/{slug}/status           : health, last run, next run
+GET    /api/apps/{slug}/runs             : execution history
+GET    /api/apps/{slug}/runs/{run_id}    : single run details
+PUT    /api/apps/{slug}/secrets/{key}    : store an API key (owner only)
+GET    /api/apps/{slug}/secrets          : list key names (owner only)
+DELETE /api/apps/{slug}/secrets/{key}    : remove a key (owner only)
+POST   /api/apps/{slug}/members          : add agent operator
+GET    /api/apps/{slug}/members          : list members
+DELETE /api/apps/{slug}/members/{agent_id} : remove member
 ```
 
 ## Examples

package/atris/skills/create-member/SKILL.md

@@ -18,7 +18,7 @@ Spec: https://github.com/atrislabs/member
 ```
 team/<name>/
 ├── MEMBER.md       REQUIRED  Persona + role + permissions
-├── SOUL.md         REQUIRED  Identity, values, lessons — who the agent is
+├── SOUL.md         REQUIRED  Identity, values, lessons: who the agent is
 ├── MISSION.md      REQUIRED  Durable purpose and goal-selection rule
 ├── goals.json      REQUIRED  Machine-readable goal state
 ├── goals.md        REQUIRED  Human-readable goal state
@@ -64,19 +64,19 @@ Use kebab-case for the name. Create the full bundle in one operation, even when
 
 ### Step 3: Write SOUL.md
 
-SOUL.md defines who the agent *is* — not what it does (MEMBER.md) or how it communicates (PERSONA.md). Use the template at `atris/team/_template/SOUL.md`.
+SOUL.md defines who the agent *is*, not what it does (MEMBER.md) or how it communicates (PERSONA.md). Use the template at `atris/team/_template/SOUL.md`.
 
 Write five sections:
 
-**Beliefs** — 3-5 convictions that guide judgment when rules don't apply. Not rules — things the agent holds true.
+**Beliefs**: 3-5 convictions that guide judgment when rules don't apply. Not rules, things the agent holds true.
 
-**Values** — 2-3 values ordered by priority. What the agent optimizes for in ambiguous situations.
+**Values**: 2-3 values ordered by priority. What the agent optimizes for in ambiguous situations.
 
-**Lessons** — Start empty. These get synthesized from the agent's journal entries over time. Not copied from elsewhere — earned from experience.
+**Lessons**: Start empty. These get synthesized from the agent's journal entries over time. Not copied from elsewhere, earned from experience.
 
-**Edges** — Honest self-assessment. What this agent is strong at and what it tends to get wrong. This helps the agent (and other agents) calibrate trust.
+**Edges**: Honest self-assessment. What this agent is strong at and what it tends to get wrong. This helps the agent (and other agents) calibrate trust.
 
-**Voice** — One sentence that captures how this agent thinks. The inner monologue, not the output style.
+**Voice**: One sentence that captures how this agent thinks. The inner monologue, not the output style.
 
 The soul should be *specific to this agent*. Two agents with the same role but different souls should make different judgment calls in ambiguous moments.
 
@@ -88,7 +88,7 @@ Use this structure:
 ---
 name: <kebab-case-name>
 role: <Human Readable Title>
-description: <one line — what this member does>
+description: <one line, what this member does>
 version: 1.0.0
 
 skills: []
@@ -100,11 +100,11 @@ permissions:
 
 Below the frontmatter, write three sections:
 
-**Persona** — How the member communicates. Tone, style, decision-making approach. Be specific — "direct and research-driven" is better than "professional and helpful."
+**Persona**: How the member communicates. Tone, style, decision-making approach. Be specific: "direct and research-driven" is better than "professional and helpful."
 
-**Workflow** — Numbered steps the member follows. This is the core operating procedure. Each step should be a concrete action.
+**Workflow**: Numbered steps the member follows. This is the core operating procedure. Each step should be a concrete action.
 
-**Rules** — Hard constraints. What the member must always or never do. Keep it to 3-5 rules.
+**Rules**: Hard constraints. What the member must always or never do. Keep it to 3-5 rules.
 
 ### Step 5: Add permissions
 

package/atris/skills/design/SKILL.md

@@ -57,7 +57,7 @@ Font alternatives: instead of Inter: Instrument Sans, Plus Jakarta Sans, Outfit.
 
 **Hierarchy:** 2-3 text levels max. don't mix 5 competing styles.
 
-**No all-caps. ever.** never use the `uppercase` tailwind class, `text-transform: uppercase`, or shout-cased copy on labels, section headers, buttons, badges, or anywhere else. random capitalized text mid-page reads as average ai-generated slop. write labels in sentence case ("Active tasks", not "ACTIVE TASKS") and let copy render as authored.
+**No all-caps. ever.** never use the `uppercase` tailwind class, the css uppercase text transform, or shout-cased copy on labels, section headers, buttons, badges, or anywhere else. random capitalized text mid-page reads as average ai-generated slop. write labels in sentence case ("Active tasks", not "ACTIVE TASKS") and let copy render as authored.
 
 **No eyebrow metadata.** Do not place tiny index, category, status, or implementation-type strips above real content. If the name changes the next action, make it a readable heading. If it does not, remove it. Generator metadata belongs in machine contracts and inspectors, never on the product canvas.
 

package/atris/skills/drive/SKILL.md

@@ -143,10 +143,10 @@ curl -s "https://api.atris.ai/api/integrations/google-drive/search?q=quarterly+r
 ```
 
 Simple queries search by file name. For advanced queries, use Drive query syntax:
-- `name contains 'budget'` — name search
-- `mimeType = 'application/vnd.google-apps.spreadsheet'` — only sheets
-- `mimeType = 'application/vnd.google-apps.document'` — only docs
-- `modifiedTime > '2026-01-01'` — recently modified
+- `name contains 'budget'`: name search
+- `mimeType = 'application/vnd.google-apps.spreadsheet'`: only sheets
+- `mimeType = 'application/vnd.google-apps.document'`: only docs
+- `modifiedTime > '2026-01-01'`: recently modified
 
 ### Get File Metadata
 ```bash
@@ -169,10 +169,10 @@ curl -s "https://api.atris.ai/api/integrations/google-drive/files/{file_id}/expo
 ```
 
 **Export formats:**
-- `text/plain` — plain text (default, good for Docs)
-- `text/html` — HTML
-- `application/pdf` — PDF
-- `text/csv` — CSV (for Sheets)
+- `text/plain`: plain text (default, good for Docs)
+- `text/html`: HTML
+- `application/pdf`: PDF
+- `text/csv`: CSV (for Sheets)
 
 ### Upload File
 ```bash
@@ -241,7 +241,7 @@ curl -s -X POST "https://api.atris.ai/api/integrations/google-drive/files/{file_
   -d '{"name": "Copy of Document", "folder_id": "OPTIONAL_FOLDER_ID"}'
 ```
 
-Works with Docs, Sheets, Slides — creates a full copy.
+Works with Docs, Sheets, Slides. It creates a full copy.
 
 ### Share File
 ```bash
@@ -440,10 +440,10 @@ curl -s "https://api.atris.ai/api/integrations/google-drive/sheets/{spreadsheet_
 ```
 
 **Range uses A1 notation:**
-- `Sheet1` — entire sheet
-- `Sheet1!A1:D10` — specific range
-- `Sheet1!A:A` — entire column A
-- `Sheet1!1:1` — entire row 1
+- `Sheet1`: entire sheet
+- `Sheet1!A1:D10`: specific range
+- `Sheet1!A:A`: entire column A
+- `Sheet1!1:1`: entire row 1
 
 ### Update Cells
 ```bash
@@ -500,20 +500,20 @@ curl -s -X POST "https://api.atris.ai/api/integrations/google-drive/sheets/{spre
 ```
 
 **Common request types:**
-- `repeatCell` — format a range (bold, colors, font size, alignment)
-- `mergeCells` — merge a range into one cell
-- `updateBorders` — add borders to cells
-- `autoResizeDimensions` — auto-fit column widths
-- `addConditionalFormatRule` — highlight cells based on rules
-- `addChart` — embed a chart
-- `addSheet` / `deleteSheet` — manage sheets/tabs
-- `sortRange` — sort rows by a column
+- `repeatCell`: format a range (bold, colors, font size, alignment)
+- `mergeCells`: merge a range into one cell
+- `updateBorders`: add borders to cells
+- `autoResizeDimensions`: auto-fit column widths
+- `addConditionalFormatRule`: highlight cells based on rules
+- `addChart`: embed a chart
+- `addSheet` / `deleteSheet`: manage sheets/tabs
+- `sortRange`: sort rows by a column
 
-**Range format:** `{"sheetId": 0, "startRowIndex": 0, "endRowIndex": 1, "startColumnIndex": 0, "endColumnIndex": 3}` — sheetId 0 = first tab, indices are 0-based.
+**Range format:** `{"sheetId": 0, "startRowIndex": 0, "endRowIndex": 1, "startColumnIndex": 0, "endColumnIndex": 3}`. Here, sheetId 0 is the first tab, and indices are 0-based.
 
 **Color format:** RGB floats 0-1, e.g. `{"red": 0.2, "green": 0.6, "blue": 1.0}`
 
-You can send multiple requests in one call — they execute in order.
+You can send multiple requests in one call. They execute in order.
 
 ---
 
@@ -571,7 +571,7 @@ You can send multiple requests in one call — they execute in order.
 ### "Format a spreadsheet"
 1. Run bootstrap
 2. Find the sheet, get its `spreadsheetId`
-3. Build requests array — bold headers, colors, borders, auto-resize, etc.
+3. Build requests array: bold headers, colors, borders, auto-resize, etc.
 4. **Show user what formatting will be applied, get approval**
 5. Apply: `POST /google-drive/sheets/{id}/batchUpdate` with `{"requests": [...]}`
 
@@ -603,7 +603,7 @@ You can send multiple requests in one call — they execute in order.
 ### "Delete files"
 1. Run bootstrap
 2. Find the files
-3. **Confirm with user** — show file names
+3. **Confirm with user**: show file names
 4. Trash (recoverable): `DELETE /google-drive/files/{id}?trash=true`
 5. Or permanent delete: `DELETE /google-drive/files/{id}`
 

package/atris/skills/email-agent/SKILL.md

@@ -119,12 +119,12 @@ curl -s "https://api.atris.ai/api/integrations/gmail/messages?query=in:inbox&max
 ```
 
 **Query syntax** (Gmail search):
-- `in:inbox` — inbox only
-- `in:inbox newer_than:1d` — today's emails
-- `is:unread` — unread only
-- `from:someone@example.com` — from specific sender
-- `subject:invoice` — subject contains word
-- `has:attachment` — emails with attachments
+- `in:inbox`: inbox only
+- `in:inbox newer_than:1d`: today's emails
+- `is:unread`: unread only
+- `from:someone@example.com`: from specific sender
+- `subject:invoice`: subject contains word
+- `has:attachment`: emails with attachments
 
 ### Read Single Email
 ```bash
@@ -158,7 +158,7 @@ curl -s -X POST "https://api.atris.ai/api/integrations/gmail/send" \
   }'
 ```
 
-**Reply in thread (IMPORTANT — use this for all replies):**
+**Reply in thread (IMPORTANT: use this for all replies):**
 
 To reply within an existing email thread, you MUST pass `thread_id` and `reply_to_message_id`. Without these, Gmail creates a new thread.
 
@@ -181,9 +181,9 @@ curl -s -X POST "https://api.atris.ai/api/integrations/gmail/send" \
   }'
 ```
 
-- `thread_id` — The thread ID from the original message. Tells Gmail which thread to add this to.
-- `reply_to_message_id` — The message ID you're replying to. The backend uses this to set `In-Reply-To` and `References` headers so Gmail threads it correctly.
-- `subject` — Must match the original subject with "Re: " prefix.
+- `thread_id`: The thread ID from the original message. Tells Gmail which thread to add this to.
+- `reply_to_message_id`: The message ID you're replying to. The backend uses this to set `In-Reply-To` and `References` headers so Gmail threads it correctly.
+- `subject`: Must match the original subject with "Re: " prefix.
 
 **With attachments:**
 ```bash
@@ -317,7 +317,7 @@ curl -s -X DELETE "https://api.atris.ai/api/integrations/gmail" \
 
 ### "Reply to this email"
 1. Run bootstrap
-2. Read the message: `GET /messages/{message_id}` — extract `id`, `thread_id`, `from`, `subject`
+2. Read the message: `GET /messages/{message_id}`, extract `id`, `thread_id`, `from`, `subject`
 3. Draft reply content
 4. **Show user the reply for approval**
 5. On approval: `POST /send` with `{to, subject: "Re: ...", body, thread_id, reply_to_message_id}`