@myapihq/cli 2.10.0 → 2.11.0

package/dist/skills/my-feedback-api/SKILL.md

@@ -0,0 +1,118 @@
+---
+name: my-feedback-api
+version: 1.0.0
+description: >
+  Collect feedback from the people using what you built. A public widget key lets a page submit without a credential; you list, filter and resolve the results. Kind is chosen by the person reporting, not inferred from their wording.
+triggers: [feedback, bug report, user feedback, feature request, widget, support, complaints, praise]
+checksum: sha256-41627c4bdf19a5af131adc0e7bab76816e7be47acdc15601bdcef7cdb47b2e26
+---
+
+# MyFeedbackAPI
+
+The loop back from the people using your product. A page collects feedback with
+a public key that authenticates nobody; you read, filter and close it.
+
+## Capabilities
+<!-- llm:start -->
+Two halves: a **widget key** a page embeds, and the **items** it produces.
+
+### The widget key is public, and that is the point
+
+`myapi feedback widget create <name>` mints a key that ships in your page
+source. It is **not a secret** — it names your org so a visitor can submit
+without signing in, and it authenticates nobody.
+
+Treating it as a credential is the mistake to avoid: people hide it, and then
+the widget cannot work. What it does need is `--origins`, so another site
+cannot post through it:
+
+```bash
+myapi feedback widget create site --origins app.example.com,example.com
+```
+
+Revoke with `myapi feedback widget revoke <id>`. Submissions stop immediately;
+feedback already collected is kept.
+
+**The browser-side submit contract is not documented yet.** The CLI and SDK
+cover minting, revoking, listing and resolving; how a page posts with only the
+widget key is not published anywhere you can read it, so don't guess the
+endpoint. Until it lands, collect through a route you control — a funnel form
+(`my-funnel-api`) or a webhook (`my-webhook-api`) — and file items with
+`myapi feedback create`.
+
+### Kind is a claim, not a guess
+
+`--kind` is one of `bug`, `issue`, `suggestion`, and it is **what the person
+reporting says it is**. Someone filing a `bug` is telling you
+they believe the product is broken — which is a different and more urgent
+signal than a wish for something new. Do not re-classify from the wording.
+
+Classification and duplicate-grouping exist in the platform's design and are
+not built yet, so treat `kind` as the person's own label rather than a
+processed signal.
+
+### Reading it back
+
+`myapi feedback list` is newest first, filterable by `--kind` and `--status`
+(`open` | `resolved`), paged with `--limit` / `--offset`. **`total` is the
+number of matches, not the size of the page**, and `has_more` flags a truncated
+result — both describe the whole match set.
+
+`myapi feedback resolve <id>` closes an item. An unknown id answers the same
+way as an already-resolved one, so a success is not proof the item existed —
+that is deliberate, so ids cannot be probed across orgs.
+<!-- llm:end -->
+
+## Commands
+<!-- generated:start -->
+| Command | What it does |
+|---|---|
+| `myapi feedback create "<text>" --kind <k>` | Record one item (`--page-url`, `--route` for context) |
+| `myapi feedback list [--kind bug\|issue\|suggestion] [--status open\|resolved] [--limit N] [--offset N]` | List feedback, newest first |
+| `myapi feedback resolve <id>` | Close an item |
+| `myapi feedback widget create <name> [--origins a.com,b.com]` | Mint a PUBLIC widget key for a site |
+| `myapi feedback widget revoke <id>` | Revoke a key; collected feedback is kept |
+<!-- generated:end -->
+
+## Examples
+<!-- llm:start -->
+```bash
+# 1. Mint a key for your site. Restrict the origins.
+myapi feedback widget create marketing --origins example.com,www.example.com
+# → prints a PUBLIC key to embed in the page
+
+# 2. Read what came back, newest first
+myapi feedback list --status open
+myapi feedback list --kind bug --limit 20
+
+# 3. Record something yourself (support call, your own testing)
+myapi feedback create "checkout 500s on the second attempt" --kind bug \
+  --route /checkout
+
+# 4. Close it
+myapi feedback resolve <id>
+```
+<!-- llm:end -->
+
+## Notes
+
+- The widget key is public by design — restrict it with `--origins` rather than hiding it.
+- `resolve` is idempotent and deliberately indistinguishable from an unknown id.
+- Feedback survives widget revocation.
+
+## HTTP (from deployed code)
+
+<!-- http:start -->
+<!-- generated by `npm run canonical-sync` — do not edit -->
+```
+base   https://api.myapihq.com
+path   POST /feedback/orgs/{org_id}/widgets
+auth   Authorization: Bearer <key>   (fn: env.__MYAPI_KEY · container: env.MYAPI_KEY)
+reply  { "success": true, "data": …, "error": null, "meta": {…} }
+```
+
+- **Per-slot host** — do not assume one host serves every slot.
+- **Org id goes in the PATH** — there is no `X-Org-Id` header.
+<!-- http:end -->
+
+Run `myapi feedback --help` for the full flag reference.

package/dist/skills/my-feedback-api/claude/.claude-plugin/plugin.json

@@ -0,0 +1,7 @@
+{
+  "name": "my-feedback-api",
+  "description": "Collect feedback from the people using what you built. A public widget key lets a page submit without a credential; you list, filter, and resolve the results.",
+  "version": "1.0.0",
+  "status": "preview",
+  "published": true
+}

package/dist/skills/my-function-api/README.md

@@ -0,0 +1,34 @@
+---
+# my-function-api
+
+Deploy JavaScript to the MyAPI edge runtime. Register a function, upload a single-file bundle, and it goes live on a public invocation URL — or runs on a cron schedule.
+
+## What it does
+
+- Two-step lifecycle: `create` registers and mints a scoped key, `deploy` ships a ≤4MB single-file bundle
+- HTTP or cron triggers, fixed at create time
+- Scoped capability key, org-locked, narrowable with `--scope <slot>,...` — a function never out-reaches the credential that made it
+- The key is injected as `env.__MYAPI_KEY` and stays current across deploys
+- Encrypted secrets via `fn env`, invocation history via `fn runs`
+
+## Quickstart
+
+```bash
+myapi fn create --name my-app-api
+myapi fn deploy fn_abc123 ./dist/bundle.js
+myapi fn runs fn_abc123
+```
+
+## Authentication
+
+```bash
+export MYAPI_KEY=hq_live_...
+```
+
+Requires `api_key` + `org_id` from **myapihq**.
+
+## Documentation
+
+Full command reference, the `__MYAPI_KEY` injection contract, and why `--scope` is effectively permanent: see `SKILL.md`.
+
+Run `myapi fn --help` for inline reference.

package/dist/skills/my-function-api/SKILL.md

@@ -4,7 +4,7 @@ version: 1.0.0
 description: >
   Deploy JavaScript functions to the MyAPI edge runtime. Register a function, upload a single-file JS bundle, get a live HTTP invocation URL or run it on a cron schedule. Each function gets a scoped capability key for cross-slot calls.
 triggers: [function, deploy function, edge function, serverless, cloudflare worker, cron, scoped api key, capability key, invocation url, bundle]
-checksum: sha256-a29b3c96645c59317c0896c9a886d9904bdb2ebde9eb8582d0cb7071d0b9ac2c
+checksum: sha256-0ab1d08fb6c9cc917fd3d1aa5d730ff3e4fe5dd37e8e30ebe225085593139cca
 ---
 
 # MyFunctionAPI
@@ -135,4 +135,19 @@ curl -H "Authorization: Bearer $SCOPED_KEY" \
 handed out — docs, front-end config, webhooks, anything given to a third
 party. Decide the full slot list before you publish the URL.
 
+## HTTP (from deployed code)
+
+<!-- http:start -->
+<!-- generated by `npm run canonical-sync` — do not edit -->
+```
+base   https://api.myapihq.com
+path   POST /function/orgs/{org_id}/functions
+auth   Authorization: Bearer <key>   (fn: env.__MYAPI_KEY · container: env.MYAPI_KEY)
+reply  { "success": true, "data": …, "error": null, "meta": {…} }
+```
+
+- **Per-slot host** — do not assume one host serves every slot.
+- **Org id goes in the PATH** — there is no `X-Org-Id` header.
+<!-- http:end -->
+
 Run `myapi fn --help` or `myapi fn <subcommand> --help` for full flag reference.

package/dist/skills/my-funnel-api/SKILL.md

@@ -4,7 +4,7 @@ version: 1.0.0
 description: >
   Create and publish websites (funnels) to the edge. Push raw HTML to any slug and it goes live instantly on your org's domain or preview subdomain.
 triggers: [funnel, landing page, website, page, publish, push, slug, html, edge, preview subdomain, makeautonomous]
-checksum: sha256-849afbb7c2c60ea88c3476d50bf51289b77cedc819a0c2b69787075ff9c23b4b
+checksum: sha256-24c12e48a6153087641369319ac67e79c1fa36b950b140621d1e118128755b53
 ---
 
 # MyFunnelAPI
@@ -121,4 +121,19 @@ myapi funnel form <funnel_id> --slug survey \
 - Deleting a funnel purges all its edge pages immediately.
 - `402 INSUFFICIENT_FUNDS` → top up (`myapi billing topup`) or enable `myapi billing auto-recharge`.
 
+## HTTP (from deployed code)
+
+<!-- http:start -->
+<!-- generated by `npm run canonical-sync` — do not edit -->
+```
+base   https://api.myfunnelapi.com
+path   POST /funnel/orgs/{org_id}/funnels
+auth   Authorization: Bearer <key>   (fn: env.__MYAPI_KEY · container: env.MYAPI_KEY)
+reply  { "success": true, "data": …, "error": null, "meta": {…} }
+```
+
+- **Per-slot host** — do not assume one host serves every slot.
+- **Org id goes in the PATH** — there is no `X-Org-Id` header.
+<!-- http:end -->
+
 Run `myapi funnel --help` or `myapi funnel <subcommand> --help` for full flag reference.

package/dist/skills/my-git-api/SKILL.md

@@ -4,7 +4,7 @@ version: 1.0.0
 description: >
   Hosted git repositories over HTTP. Create repos, read history (log/show/tree/blob/diff), and write atomically (commit/branch/tag/merge) — no clone needed. Real `git clone`/`push` also work over HTTPS with your API key as the password.
 triggers: [git, repo, repository, clone, push, commit, branch, tag, merge, diff, version control, source control, vcs]
-checksum: sha256-417bafc3e56707f637937f98341d6200c8a9bb6ad5d0b13f84f7b6810e0340f5
+checksum: sha256-0ce9ebdde1ec29160d186ab0e4fb254d47473af3b57829f0156845bfb08c4ab1
 ---
 
 # MyGitAPI
@@ -117,6 +117,7 @@ surface never leaks which repos exist).
 - **`commit --base`** is the concurrency guard: pass the expected tip SHA to reject a stale write, or `--base ""` to require the branch be newly created. Omit it to create-or-update.
 - **Metering.** A `git push` (and `myapi git commit`) is metered like a commit. Clone/fetch (`upload-pack`) is free.
 - **Limits.** A push body is capped at 100 MiB; per-repo size limits apply (a push past the cap is reported as a receive-pack failure).
+- **`402`** — `INSUFFICIENT_FUNDS`: top up or enable `myapi billing auto-recharge`. `SPEND_CAP_EXCEEDED`: raise your own ceiling with `myapi billing spend-cap`.
 - **Auth field.** git may put the API key in the username or password slot — both work. Embedding it in the URL (`https://x:$KEY@…`) avoids the interactive prompt but writes the API key into `.git/config`; use a credential helper for anything persistent.
 
 ## Commit authorship
@@ -125,4 +126,19 @@ surface never leaks which repos exist).
 author. Without them the commit is attributed to the API key's account, which
 makes every agent-written commit look like the same person.
 
+## HTTP (from deployed code)
+
+<!-- http:start -->
+<!-- generated by `npm run canonical-sync` — do not edit -->
+```
+base   https://api.myapihq.com
+path   POST /git/orgs/{org_id}/repos
+auth   Authorization: Bearer <key>   (fn: env.__MYAPI_KEY · container: env.MYAPI_KEY)
+reply  { "success": true, "data": …, "error": null, "meta": {…} }
+```
+
+- **Per-slot host** — do not assume one host serves every slot.
+- **Org id goes in the PATH** — there is no `X-Org-Id` header.
+<!-- http:end -->
+
 Run `myapi git --help` for the full flag reference.

package/dist/skills/my-image-api/SKILL.md

@@ -4,7 +4,7 @@ version: 1.0.0
 description: >
   Generate AI images from a text prompt. Async — submit a prompt, the CLI polls until the image is ready, then returns a public CDN URL. Image lands automatically in your org's storage.
 triggers: [image, generate image, ai image, hero image, logo, illustration, watercolor, prompt, cyberpunk, ratio, aspect ratio]
-checksum: sha256-17f496ae8c575743872e87c5c3b281e621b3557ef3ff9eeb0b3d664ec4212e57
+checksum: sha256-c12686cd9f53f18dbc692c3aca9391409cc1540b693b6f2059074e02bafff72c
 ---
 
 # MyImageAPI
@@ -73,6 +73,22 @@ Tips for agents:
 
 - Cost is per-image at the selected model's catalog rate (`myapi image models`). Failed jobs aren't charged.
 - The asset URL is public — don't generate sensitive content.
+- **`402`** — `INSUFFICIENT_FUNDS`: top up or enable `myapi billing auto-recharge`. `SPEND_CAP_EXCEEDED`: raise your own ceiling with `myapi billing spend-cap`.
 - `delete` is permanent for the asset; the job record (prompt + metadata) stays for history. Look before you delete: `myapi image list` first, pass `--org` explicitly; delete verbs require `--yes` in non-interactive runs.
 
+## HTTP (from deployed code)
+
+<!-- http:start -->
+<!-- generated by `npm run canonical-sync` — do not edit -->
+```
+base   https://api.myimageapi.com
+path   POST /image/orgs/{org_id}/generate
+auth   Authorization: Bearer <key>   (fn: env.__MYAPI_KEY · container: env.MYAPI_KEY)
+reply  { "success": true, "data": …, "error": null, "meta": {…} }
+```
+
+- **Per-slot host** — do not assume one host serves every slot.
+- **Org id goes in the PATH** — there is no `X-Org-Id` header.
+<!-- http:end -->
+
 Run `myapi image --help` or `myapi image <subcommand> --help` for full flag reference.

package/dist/skills/my-llm-api/SKILL.md

@@ -7,7 +7,7 @@ description: >
   (classify / extract / summarize / draft) that hide the model behind a
   task. Pricing in cents per 1M tokens; charged from your MyAPI balance.
 triggers: [llm, completion, chat, embed, embedding, inference, classify, extract, summarize, draft, qwen]
-checksum: sha256-2a9070e118134aea98a1be986d29f9c4e3a969aa960e7dfc87137bbc490ac04e
+checksum: sha256-d89e5c7ccadb7e967f97ec201b745f2313e4263f8f08a695f4cc9fbaa0c4b500
 ---
 
 # MyLLMAPI
@@ -17,8 +17,6 @@ A two-surface LLM gateway:
 - **Raw** — `complete` / `embed` / `models`. You pick a self-hosted catalog model (run `myapi llm models` for the live list). Proprietary models are not callable here.
 - **Verbs** — `classify` / `extract` / `summarize` / `draft`. You ask for a task done; the model is implementation detail.
 
-Pricing is cents per 1M tokens at the actual upstream rate, debited from your MyAPI balance.
-
 ## Capabilities
 <!-- llm:start -->
 For workflow tasks — summarize, classify, extract, draft. The reply goes to stdout and the usage footer (tokens, cost, finish reason) to stderr, so `myapi llm complete ... | jq` works.
@@ -112,7 +110,7 @@ r = client.chat.completions.create(model="<model-id>",
 | `myapi llm classify "<input>" --labels <csv> [--multi] [--tier <t>] [--json]` | Pick a label from a set |
 | `myapi llm extract "<input>" --schema <path\|json> [--tier <t>] [--json]` | Pull structured data conforming to a JSON Schema |
 | `myapi llm summarize "<input>" [--style brief\|exec\|bullet] [--tier <t>] [--json]` | Summarize text |
-| `myapi llm draft --kind <what> [--prompt "<s>"] [--context <json>] ["<src>"] [--tier <t>] [--json]` | Draft an email / reply / message / … |
+| `myapi llm draft --kind <what> [--prompt "<s>"] [--facts <json>] [--directives <json>] ["<src>"] [--tier <t>] [--json]` | Draft an email / reply / message / … |
 <!-- generated:end -->
 
 Pass `-` as the prompt/input to read from stdin. Pass `--file <path>` to read longer content from disk.
@@ -145,7 +143,7 @@ myapi llm summarize --file long-thread.txt --style bullet
 
 myapi llm draft --kind email \
   --prompt "Friendly welcome, under 60 words." \
-  --context '{"recipient":"a new signup","product":"MyAPI"}'
+  --facts '{"recipient":"a new signup","product":"MyAPI"}'
 
 # Classify + route an inbound webhook delivery
 BODY=$(myapi webhook delivery <id> --json | jq -r '.body')
@@ -156,7 +154,8 @@ INTENT=$(printf '%s' "$BODY" | myapi llm classify - \
 
 ## Notes
 
-- **`draft` context safety.** `context` fields are quoted into the prompt verbatim; sensitive-named keys (`secret`, `api_key`, `password`, …) are NOT redacted. Two guards on top: (a) injection-defense strips `instructions`/`system`/`prompt`/`override` keys and surfaces them in `meta.warnings`; (b) output guardrail substring-scans fact values (length ≥ 4) in the response and lists matches in `meta.guardrails.facts_in_output` (signal, not redaction). Rule of thumb: never put credentials, PII, or internal metadata in `context` — pass identifiers, reference them indirectly.
+- **`draft --facts` safety.** Fact values are quoted into the prompt verbatim and sensitive-named keys (`secret`, `api_key`, `password`, …) are NOT redacted. Two guards: injection-defense strips `instructions`/`system`/`prompt`/`override` keys into `meta.warnings`; an output guardrail substring-scans fact values (≥4 chars) and lists hits in `meta.guardrails.facts_in_output` (signal, not redaction). Never put credentials, PII, or internal metadata in `--facts` — pass identifiers and reference them indirectly.
+- **`402`** — `INSUFFICIENT_FUNDS`: top up or enable `myapi billing auto-recharge`. `SPEND_CAP_EXCEEDED`: raise your own ceiling with `myapi billing spend-cap`.
 - **Self-hosted raw, server-picked verbs.** Raw runs on MyAPI's TPU; verbs route wherever the server picks.
 - **Cost + latency.** `usage.cost_cents` is authoritative — no markup. Varies by tier: 200–600 ms to first token, 1–3 s end-to-end.
 - **Live catalog, no streaming, no BYOK.** Don't hard-code ids — `models` is truth (CLI auto-picks if `--model` omitted). Full reply only.
@@ -173,3 +172,18 @@ myapi llm draft --kind email --prompt "the invoice is due" \
 ```
 
 `--context` is the old name for `--facts`; accepted, deprecated upstream.
+
+## HTTP (from deployed code)
+
+<!-- http:start -->
+<!-- generated by `npm run canonical-sync` — do not edit -->
+```
+base   https://api.myapihq.com
+path   POST /llm/orgs/{org_id}/complete
+auth   Authorization: Bearer <key>   (fn: env.__MYAPI_KEY · container: env.MYAPI_KEY)
+reply  { "success": true, "data": …, "error": null, "meta": {…} }
+```
+
+- **Per-slot host** — do not assume one host serves every slot.
+- **Org id goes in the PATH** — there is no `X-Org-Id` header.
+<!-- http:end -->

package/dist/skills/my-payments-api/README.md

@@ -0,0 +1,33 @@
+---
+# my-payments-api
+
+Take payments via Stripe. Connect your own Stripe account, then create hosted Checkout URLs for one-off or recurring charges. No raw card data touches your code.
+
+## What it does
+
+- Bring your own Stripe key (T0) — your account stays the merchant of record
+- One verb for both models: `charge --amount 9.99`, or add `--every month|year` for a subscription
+- Returns a hosted Stripe Checkout URL to paste into a frontend, email, or redirect
+- Full refunds; list and inspect charges
+
+## Quickstart
+
+```bash
+myapi payments connect --stripe-key sk_live_...
+myapi payments charge --amount 9.99 --description "Pro plan upgrade"
+myapi payments refund <charge_id>
+```
+
+## Authentication
+
+```bash
+export MYAPI_KEY=hq_live_...
+```
+
+Requires `api_key` + `org_id` from **myapihq**, plus your own Stripe secret key.
+
+## Documentation
+
+Full command reference and what is deliberately not available yet (T1 Connect Express, partial refunds, coupons, Stripe Tax): see `SKILL.md`.
+
+Run `myapi payments --help` for inline reference.

package/dist/skills/my-payments-api/SKILL.md

@@ -4,7 +4,7 @@ version: 1.1.0
 description: >
   Take payments via Stripe — one-off or recurring. Connect your own Stripe account, then create hosted Stripe Checkout URLs. No raw card data ever touches your code.
 triggers: [payment, stripe, checkout, charge, subscription, take payments, billing, refund, recurring, sell]
-checksum: sha256-bdf124b2a3d32c87676ad34f0a20df134099bcbebe7f9a11f05ceddf74a6b2a9
+checksum: sha256-742d42da2441fb3ce1317b503dd68f706a1f3d4a7d0f8128473683c26fa3be18
 ---
 
 # MyPaymentsAPI
@@ -82,4 +82,19 @@ myapi payments refund <charge_id>
 - Multi-currency on a single charge.
 - Customer object as a first-class resource (Stripe tracks by email; `my-crm-api` has the engagement view).
 
+## HTTP (from deployed code)
+
+<!-- http:start -->
+<!-- generated by `npm run canonical-sync` — do not edit -->
+```
+base   https://api.myapihq.com
+path   POST /payments/orgs/{org_id}/connect
+auth   Authorization: Bearer <key>   (fn: env.__MYAPI_KEY · container: env.MYAPI_KEY)
+reply  { "success": true, "data": …, "error": null, "meta": {…} }
+```
+
+- **Per-slot host** — do not assume one host serves every slot.
+- **Org id goes in the PATH** — there is no `X-Org-Id` header.
+<!-- http:end -->
+
 Run `myapi payments --help` or `myapi payments <subcommand> --help` for full flag reference.

package/dist/skills/my-people-api/SKILL.md

@@ -4,7 +4,7 @@ version: 1.0.0
 description: >
   Contact database backed by the Goldfox crawl. Filter people by Goldfox confidence tier, seniority, email type, country, link confidence, plus rich behavioral company signals (has_c_level, has_careers_page, has_decision_maker, etc.). The targeting layer for outbound campaigns.
 triggers: [people, contacts, leads, prospects, search, filter, goldfox, decision-makers, c-level, b2b targeting, corporate email]
-checksum: sha256-1eb93087638966eee65fb7a88fe37baaa27a01dd5d97c1ef17387b1a36300c0e
+checksum: sha256-e3f793369aebacfedad165f6bfa1ef831cbf17951e18f34110197697149bf416
 ---
 
 # MyPeopleAPI
@@ -104,4 +104,19 @@ myapi audience members $AID --limit 100 --json > targets.json
 - `keyword` matches the row's **domain**, not name/title. Use `--keyword acme` to find people whose domain contains "acme".
 - For a persistent target list, use `my-audience-api` (snapshot the filter; re-evaluate on `audience refresh`).
 
+## HTTP (from deployed code)
+
+<!-- http:start -->
+<!-- generated by `npm run canonical-sync` — do not edit -->
+```
+base   https://api.myapihq.com
+path   POST /people/orgs/{org_id}/search
+auth   Authorization: Bearer <key>   (fn: env.__MYAPI_KEY · container: env.MYAPI_KEY)
+reply  { "success": true, "data": …, "error": null, "meta": {…} }
+```
+
+- **Per-slot host** — do not assume one host serves every slot.
+- **Org id goes in the PATH** — there is no `X-Org-Id` header.
+<!-- http:end -->
+
 Run `myapi people --help` for full flag reference.

package/dist/skills/my-pixel-api/SKILL.md

@@ -4,7 +4,7 @@ version: 1.0.0
 description: >
   Tracking pixel + identity resolution for MyAPI funnels and email. Capture visits and events, resolve known users to anonymous sessions, stream interaction events for analytics. Pairs with mycrmapi for auto-ingest of pixel_visit events on known contacts.
 triggers: [pixel, analytics, tracking, visit, event, identity, session, attribution, geo, open pixel]
-checksum: sha256-562f081ef28c0dae01045a4135f9d073d69c50899145e75c6c97fe75b974cb15
+checksum: sha256-2c6508a5fdfa752cc5fd4231c087380e6d904f514973830a34e919b277ad31de
 ---
 
 # MyPixelAPI
@@ -50,9 +50,13 @@ Pixel data lives ~90 days hot; aggregated geo/audience samples persist longer. D
 | `myapi pixel events (--campaign-id <id> \| --domain <domain>) [--from <iso>] [--to <iso>] [--limit N] [--offset N]` | Engagement events (sent / open / click / page_visit). Requires at least one filter: `--campaign-id` or `--domain` |
 | `myapi pixel interactions [--website <domain>] [--campaign-id <id>] [--domain <domain>] [--from <iso>] [--to <iso>] [--limit N] [--offset N]` | Unified timeline of visits + events |
 | `myapi pixel identity <pixel_id> --website <domain>` | Resolve a pixel ID's identity graph (returns `{uuid, is_resolved, nodes, …}`) |
+| `myapi pixel identify <pixel_id> (--email <addr> \| --external-id <id>)` | Attach a known identity to an anonymous visitor — the write half of identity resolution |
 | `myapi pixel audience` | Geographic distribution sample of your pixel audience |
 <!-- generated:end -->
 
+`identity` reads the graph; `identify` writes to it. Everything else in this
+slot is read-only.
+
 ## Examples
 <!-- llm:start -->
 ```bash
@@ -94,8 +98,27 @@ visitor:
 Pass either or both. After this, `pixel identity <pixel_id>` resolves the
 graph across the visitor's sessions.
 
+## HTTP (from deployed code)
+
+<!-- http:start -->
+<!-- generated by `npm run canonical-sync` — do not edit -->
+```
+base   https://api.mypixelapi.com
+path   POST /pixel/orgs/{org_id}/identify
+auth   Authorization: Bearer <key>   (fn: env.__MYAPI_KEY · container: env.MYAPI_KEY)
+reply  { "success": true, "data": …, "error": null, "meta": {…} }
+```
+
+- **Per-slot host** — do not assume one host serves every slot.
+- **Org id goes in the PATH** — there is no `X-Org-Id` header.
+<!-- http:end -->
+
 Run `myapi pixel --help` for inline reference.
 
-## Status
+## Maturity
 
-**Backend reads are GA; agent-facing skill content is draft.** This skill publishes (status flips from `planned` to `ga` in `services.ts`) once pixel → CRM auto-ingest is wired so the cross-primitive examples above resolve end-to-end.
+Reads are live and the CLI surface is stable. The slot is `preview` for one
+reason: **pixel → CRM auto-ingest is not wired yet**, so a `pixel_visit` does
+not appear on a contact's CRM timeline today. Every read verb works; the
+cross-primitive examples that join pixel data to CRM contacts do not resolve
+end-to-end until that lands.

package/dist/skills/my-queue-api/SKILL.md

@@ -4,7 +4,7 @@ version: 1.0.0
 description: >
   Durable job queue — enqueue work and have it retried against your HTTP consumer, with concurrency caps and a dependency DAG.
 triggers: [queue, job queue, background job, enqueue, retry, async work, dead letter, delayed job, concurrency]
-checksum: sha256-528a48bc9adcde7c48ad7208db7796ec37032f6df31bf5a2ca1f2bf63f643fe9
+checksum: sha256-e3079f05a9f0b6eb1860749e42a3feea3cc4f18997b8bd6a0d5ff892e29bc628
 ---
 
 # MyQueueAPI
@@ -29,7 +29,7 @@ A job carries an arbitrary JSON `payload` (≤256 KB). Jobs support:
 - `delay_seconds` — hold the job before it becomes eligible.
 - `depends_on` — job ids this job waits on. Immutable — a DAG by construction. A job with unsucceeded `depends_on` starts blocked.
 
-**queue vs workflow vs task** — see `docs/orchestration-decision-guide.md`. Short version: `workflow` reacts to inbound webhooks inline; `queue` runs durable retried machine work; `task` is for work that needs an agent/human decision.
+**queue vs workflow vs task** — see the section at the end of this skill.
 <!-- llm:end -->
 
 ## Commands
@@ -75,4 +75,37 @@ myapi queue job <job_id>
 - `depends_on` is immutable: the DAG is declared at enqueue time.
 - `payload` is capped at 256 KB.
 
+## HTTP (from deployed code)
+
+<!-- http:start -->
+<!-- generated by `npm run canonical-sync` — do not edit -->
+```
+base   https://api.myapihq.com
+path   POST /queue/orgs/{org_id}/queues
+auth   Authorization: Bearer <key>   (fn: env.__MYAPI_KEY · container: env.MYAPI_KEY)
+reply  { "success": true, "data": …, "error": null, "meta": {…} }
+```
+
+- **Per-slot host** — do not assume one host serves every slot.
+- **Org id goes in the PATH** — there is no `X-Org-Id` header.
+<!-- http:end -->
+
+## workflow vs queue vs task
+
+| If you are saying… | Use |
+|---|---|
+| "When this webhook arrives, do these steps." | **workflow** |
+| "Run this reliably, eventually — retry it, don't block on it." | **queue** |
+| "Someone (agent or human) needs to look at this and decide." | **task** |
+
+**Retry is not the distinction** — workflow and queue both retry. What differs
+is where the work runs: a workflow step runs inline during the delivery that
+triggered it; a queue job runs asynchronously against your external
+`consumer_url`, bounded by `max_concurrency`.
+
+They chain one way: a workflow's `enqueue_job` step hands durable work to a
+queue. And they listen to two different surfaces — workflow fires on inbound
+**webhook deliveries**; task `resolve_on` matches the internal **platform
+events** bus, which is not the CRM timeline.
+
 Run `myapi queue --help` for the full flag reference.