@dsalazardev/ai-hub-cli 0.0.0-stage → 0.1.0

package/skills/shared/coolify/scripts/verify.sh

@@ -0,0 +1,123 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+# verify.sh — coolify skill gate. Run from your PROJECT root (or this skill's dir).
+#
+# Static, read-only lint of any committed docker-compose artifact (docker-compose.yml/.yaml,
+# compose.yml/.yaml) plus a check that this skill's SKILL.md documents the canonical port matrix.
+# It NEVER writes, NEVER connects to the network, and NEVER deploys.
+#
+# Per compose file it asserts:
+#   (a) no hardcoded secret literal — values for *PASSWORD*/*SECRET*/*TOKEN*/*_KEY*/DATABASE_URL
+#       must be env-refs (${...}) , not inline literals.
+#   (b) any database service (postgres/mysql/mariadb/mongo) is backed by a NAMED volume (data-loss guard).
+#   (c) at least one healthcheck: present.
+#   (d) no floating ":latest" image tag, and no image without an explicit tag (also "floating").
+# And once: SKILL.md mentions every canonical port (8000 80 443 6001 6002).
+#
+# Exit code: non-zero ONLY on a hard failure (a/b/c/d or a missing port). An EMPTY/CLEAN target
+# (no compose files found) exits 0 with a skip — never a false failure. Missing optional context
+# (no SKILL.md to check) is advisory, not fatal.
+#
+# Portability: stock macOS bash 3.2 (no mapfile, no associative arrays). Arrays are initialised so
+# they expand safely under `set -u`.
+
+YELLOW=$'\033[33m'; GREEN=$'\033[32m'; RED=$'\033[31m'; NC=$'\033[0m'
+EXIT=0
+skip() { printf '%s[skip]%s %s\n' "$YELLOW" "$NC" "$*"; }
+warn() { printf '%s[warn]%s %s\n' "$YELLOW" "$NC" "$*"; }
+ok()   { printf '%s[ok]%s %s\n'   "$GREEN"  "$NC" "$*"; }
+err()  { printf '%s[fail]%s %s\n' "$RED"    "$NC" "$*"; EXIT=1; }
+
+ROOT="$(pwd)"
+SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"
+
+# --- discover compose files (skip vendor dirs) ---
+COMPOSE_FILES=()
+while IFS= read -r -d '' f; do
+  COMPOSE_FILES+=("$f")
+done < <(
+  find "$ROOT" \
+    \( -path '*/node_modules/*' -o -path '*/.git/*' -o -path '*/vendor/*' -o -path '*/.venv/*' \) -prune -o \
+    -type f \( -name 'docker-compose.yml' -o -name 'docker-compose.yaml' \
+               -o -name 'compose.yml' -o -name 'compose.yaml' \) -print0 2>/dev/null
+)
+
+lint_compose() {
+  file="$1"
+  base="$(basename "$file")"
+
+  # (a) hardcoded secrets: a *PASSWORD/SECRET/TOKEN/_KEY/DATABASE_URL key whose value is NOT a ${...} ref
+  #     and is not empty. We look at "key: value" pairs (env blocks are key: value in compose).
+  if grep -Eiq '^[[:space:]]*[A-Za-z_]*(password|secret|token|_key|database_url)[A-Za-z_]*[[:space:]]*:[[:space:]]*[^$[:space:]"'"'"'].*' "$file"; then
+    # exclude the case where the value is an env-ref like ${FOO}
+    if grep -Ei '^[[:space:]]*[A-Za-z_]*(password|secret|token|_key|database_url)[A-Za-z_]*[[:space:]]*:[[:space:]]*' "$file" \
+       | grep -Eiv ':[[:space:]]*("?\$\{|\$[A-Za-z_])' \
+       | grep -Eq ':[[:space:]]*[^[:space:]$]'; then
+      err "$base: hardcoded secret literal — use an env-ref (\${VAR}) injected by Coolify, not an inline value"
+    fi
+  fi
+
+  # (b) named volume for any database service. If a db image is present, require a top-level `volumes:` map.
+  if grep -Eiq 'image:[[:space:]]*("?)(postgres|mysql|mariadb|mongo)' "$file"; then
+    if grep -Eq '^volumes:' "$file"; then
+      ok "$base: database service has a named volume declaration"
+    else
+      err "$base: database service without a named volume — data is wiped on recreate"
+    fi
+  fi
+
+  # (c) at least one healthcheck
+  if grep -Eq '^[[:space:]]*healthcheck:' "$file"; then
+    ok "$base: healthcheck present"
+  else
+    err "$base: no healthcheck: — Coolify/Traefik can't tell when the container is ready"
+  fi
+
+  # (d) floating tags: image: foo:latest, or image: foo with no tag at all
+  while IFS= read -r line; do
+    img="$(printf '%s' "$line" | sed -E 's/^[[:space:]]*image:[[:space:]]*"?//; s/"?[[:space:]]*$//')"
+    case "$img" in
+      \$*) : ;;                                   # env-ref image, skip
+      *:latest) err "$base: floating ':latest' tag on image '$img' — pin a tag or digest" ;;
+      *@sha256:*) : ;;                            # digest-pinned, fine
+      *:*) : ;;                                   # has an explicit tag, fine
+      *) err "$base: image '$img' has no tag (floating) — pin a tag or digest" ;;
+    esac
+  done < <(grep -E '^[[:space:]]*image:[[:space:]]*' "$file" || true)
+}
+
+if [ "${#COMPOSE_FILES[@]}" -eq 0 ]; then
+  skip "no docker-compose files found — nothing to lint (clean)"
+else
+  for f in "${COMPOSE_FILES[@]}"; do
+    lint_compose "$f"
+  done
+fi
+
+# --- port matrix documented in SKILL.md (look next to this script, then under ROOT) ---
+SKILL_MD=""
+if [ -f "$SCRIPT_DIR/../SKILL.md" ]; then
+  SKILL_MD="$SCRIPT_DIR/../SKILL.md"
+else
+  cand="$(find "$ROOT" -type f -name 'SKILL.md' -path '*coolify*' -print 2>/dev/null | head -n1 || true)"
+  [ -n "$cand" ] && SKILL_MD="$cand"
+fi
+
+if [ -n "$SKILL_MD" ] && [ -f "$SKILL_MD" ]; then
+  missing=""
+  for p in 8000 80 443 6001 6002; do
+    grep -Eq "(^|[^0-9])$p([^0-9]|$)" "$SKILL_MD" || missing="$missing $p"
+  done
+  if [ -n "$missing" ]; then
+    err "SKILL.md missing canonical port(s):$missing"
+  else
+    ok "SKILL.md documents the canonical port matrix (8000/80/443/6001/6002)"
+  fi
+else
+  skip "no coolify SKILL.md found to check the port matrix"
+fi
+
+printf '\n'
+if [ "$EXIT" -eq 0 ]; then ok "verify.sh passed"; else err "verify.sh found failures"; fi
+exit "$EXIT"

package/skills/shared/cost-tracking/SKILL.md

@@ -0,0 +1,176 @@
+---
+name: cost-tracking
+description: "Use when metering and capping AI or cloud app spend — tokens read from the response `usage` object, priced off a dated rate table, ledgered per user/tenant/feature, with alerts and a hard cap before the bill. NOT cash runway (that is `finance-ops`), NOT cost-per-unit margin (that is `unit-economics`), NOT booking the spend (that is `bookkeeping`)."
+tags: ["cost-tracking", "token-accounting", "llm-spend", "budgets", "alerts", "prompt-caching", "billing-guardrails", "finops"]
+recommends: ["finance-ops", "unit-economics", "analytics", "stripe", "aws-essentials", "gcp-essentials"]
+origin: ai-hub
+---
+
+# Cost tracking
+
+You are building the money meter for an AI or cloud app: every model call gets a token count, a price, a ledger row, and a budget check — so a cap fires *before* the invoice, not when finance forwards it in a panic. Model-API spend roughly doubled from $3.5B to $8.4B between late 2024 and mid 2025 (firecrawl.dev best-llm-observability-tools, accessed 2026-06-02); the bill is now big enough to need a guardrail, not a spreadsheet at month-end.
+
+The chain you build, in order — each step feeds the next:
+
+1. **Capture** — read tokens from the response, not from a pre-send guess.
+2. **Price** — multiply tokens by a *versioned, dated* rate table.
+3. **Ledger** — append one idempotent row per request, tagged for attribution.
+4. **Budget** — roll the ledger up against a soft and a hard threshold.
+5. **Guard** — alert, degrade, or refuse before the threshold becomes an invoice.
+
+The one rule that organizes everything: **bill against the response `usage` object. Anything you compute before the call is an estimate — good only for the pre-flight cap check, never for the ledger.**
+
+## What this skill produces
+
+A checkable cost setup: a **pricing table** (each model row carries `effective_date` + `source`), an **append-only ledger** schema (idempotency key + attribution keys), and a **budget** with both a soft and a hard threshold. `scripts/verify.sh` lints those artifacts (last section). Prose alone is not a deliverable — emit the config.
+
+## Capture: read `usage`, do not estimate
+
+Pre-send token counts (`tiktoken`, Anthropic `client.messages.count_tokens()`) are estimates. They exist for the *pre-flight cap check* — "will this request likely breach the budget?" — and nothing else. The truth lands in the response: input, output, **cached**, and **reasoning** tokens, plus audio/image tokens where the modality applies. Bill the ledger off that object.
+
+Two provider facts that bite if you assume otherwise:
+- Anthropic `count_tokens()` returns *input* tokens only, is free, has its own rate limit, and is still an estimate. Anthropic is **not** tiktoken-compatible — do not reuse an OpenAI tokenizer to price Claude (platform.claude.com token-counting; github.com/anthropics/anthropic-tokenizer-typescript, accessed 2026-06-02).
+- Output and reasoning tokens are the expensive half — output runs 4-5x the input rate. A meter that only counts input is wrong by most of the bill.
+
+```python
+# Bad: pricing off a pre-send character/word guess. Wrong, and ignores output.
+est_tokens = len(prompt) // 4
+cost = est_tokens * rate_in            # output + reasoning never counted
+
+# Good: capture every field the response actually reports, then price that.
+resp = client.messages.create(model=model, messages=msgs, max_tokens=1024)
+u = resp.usage
+record = {
+    "input_tokens":  u.input_tokens,
+    "output_tokens": u.output_tokens,
+    "cache_read_tokens":     getattr(u, "cache_read_input_tokens", 0),
+    "cache_write_tokens":    getattr(u, "cache_creation_input_tokens", 0),
+    # OpenAI exposes cached as usage.prompt_tokens_details.cached_tokens
+}
+cost = price(model, record)            # see "pricing is data" below
+```
+
+Wrap the SDK call **once** so capture cannot be skipped. A meter you have to remember to call is a meter that's already missing rows (langfuse.com token-and-cost-tracking, accessed 2026-06-02).
+
+## Pricing is data, never literals
+
+Rates drift fast and silently mis-bill when stale. Keep prices in a **versioned table** — one row per model, each with `effective_date` and `source` — loaded as data. Never write a rate as a literal in business logic. Look up by model and **fail loud on an unknown model; never default to $0**, or a new model silently bills as free and the leak is invisible.
+
+```yaml
+# pricing.yaml — perishable. Verify against source before trusting. Dated 2026-06-02.
+models:
+  - model: claude-haiku-4.5
+    effective_date: 2026-06-02
+    source: cloudzero.com/blog/claude-api-pricing
+    input_per_mtok: 1.00
+    output_per_mtok: 5.00
+    cache_read_per_mtok: 0.10
+  - model: claude-sonnet-4.6
+    effective_date: 2026-06-02
+    source: cloudzero.com/blog/claude-api-pricing
+    input_per_mtok: 3.00
+    output_per_mtok: 15.00
+    cache_read_per_mtok: 0.30
+  - model: claude-opus-4.7
+    effective_date: 2026-06-02
+    source: cloudzero.com/blog/claude-api-pricing
+    input_per_mtok: 5.00
+    output_per_mtok: 25.00
+    cache_read_per_mtok: 0.50
+  - model: gpt-5.5
+    effective_date: 2026-06-02
+    source: openai.com/api/pricing
+    input_per_mtok: 5.00
+    output_per_mtok: 40.00
+    cache_read_per_mtok: 0.50   # OpenAI cached input = 90% off standard input
+```
+
+These numbers are a **snapshot, not a constant** — model names and rates move month to month. Dated 2026-06-02 from the sources above. The dated per-provider snapshots, the `usage`-field map per provider, and refresh instructions live in `references/pricing-tables.md`; read it before you trust a rate.
+
+## Ledger: append-only, idempotent, attributed
+
+One row per request, append-only. Two things must be on every row or the ledger lies:
+
+- A **request/idempotency key.** Retries and SDK auto-retries fire the same logical call twice; without a key the row is written twice and you double-count spend.
+- At least one **attribution key** (user, tenant, feature, model). A per-org total can tell you the bill is high; it cannot tell you *which feature or customer* is the leak. Attribution is the difference between "spend is up" and "the summarize-document feature on the enterprise tenant tripled."
+
+```sql
+-- append-only; (request_id) is the idempotency key — upsert, never plain insert
+CREATE TABLE llm_cost_ledger (
+  request_id     TEXT PRIMARY KEY,           -- idempotency: retries collapse to one row
+  ts             TIMESTAMPTZ NOT NULL,
+  model          TEXT NOT NULL,
+  input_tokens   INTEGER NOT NULL,
+  output_tokens  INTEGER NOT NULL,
+  cached_tokens  INTEGER NOT NULL DEFAULT 0,
+  cost_usd       NUMERIC(12,6) NOT NULL,      -- priced from the table above
+  user_id        TEXT,                        -- attribution keys
+  tenant_id      TEXT,
+  feature        TEXT
+);
+```
+
+This is an *operational* ledger, not the accounting record — categorizing the spend into the books is `bookkeeping`, and the same rows feed the cost numerator in `unit-economics` and one input line in `finance-ops`. "Cost per active user" on the behavior side is `analytics`; charging customers for metered usage is `stripe`.
+
+## Budgets, alerts, caps
+
+A budget needs a **soft** state (alert + degrade) and a **hard** state (refuse). Roll the ledger up per window (day/month) and per attribution key, then branch:
+
+| Spend vs budget | State | Action |
+|---|---|---|
+| < 50% | normal | log only |
+| 50% / 80% | warn | fire alert to the same pipe as cloud alerts; no behavior change |
+| 100% | **soft cap** | degrade — downshift to a cheaper model, drop optional/enrichment calls, shrink context |
+| over hard cap | **hard cap** | refuse the request with a typed error (`BudgetExceededError`), not a silent failure |
+
+Two distinct checks, do not conflate them:
+- **Pre-flight gate** uses the *estimate* (pre-send token count × rate) to refuse a request that would obviously blow the hard cap — this is the only legitimate use of an estimate.
+- **Post-hoc reconciliation** rolls up the *ledger* (real `usage`) against the budget. When someone asks "why is the bill 3x the estimate," you compare provider-billed usage to your ledgered `usage` — the gap is almost always uncounted output/reasoning/cache-write tokens or missing rows from un-wrapped call sites.
+
+## The three levers that move the number
+
+Don't assert savings — show the break-even. Pricing per fact-checked sources accessed 2026-06-02 (platform.claude.com prompt-caching; finout.io anthropic-api-pricing).
+
+- **Prompt caching.** Cache *reads* cost 0.1x base input. The 5-min write costs 1.25x, the 1-hour write 2x. So a 5-min entry pays for itself after roughly **one** cache hit: `1.25 + 0.1·h` (cached) beats `1·(1+h)` (uncached) once `h ≥ 1`. Caching a stable system prompt across a session is almost always net cheaper.
+- **Batch API.** 50% off on both OpenAI and Anthropic, async within 24h, and it **stacks with caching** — combined up to ~95% off. Use it for anything not user-facing-realtime: evals, backfills, nightly summaries.
+- **Model downshift.** The largest lever. Route easy requests to Haiku/cheap models and reserve Opus/GPT-5.5 for hard ones — at 5x the rate, downshifting the routable half of traffic dwarfs a few percent of caching.
+
+"We'll add caching later" without measuring the hit rate is a guess, not a lever. Instrument `cache_read_tokens` in the ledger first, then you know.
+
+## Cloud spend is the slow backstop
+
+App-level metering is your real-time guard. Cloud billing alerts are a **delayed backstop** — useful, but never the thing standing between you and a runaway loop.
+
+- **AWS** Cost Anomaly Detection is ML-based, runs ~3x/day with up to 24h data delay → SNS → Lambda. AWS Budgets adds threshold alerts on the same pipe.
+- **GCP/Azure** are threshold-only. GCP's budget → Pub/Sub → Cloud Function is the one that can actually *pause or throttle* a workload programmatically.
+
+Route every cloud alert into the **same alert pipe** as your app-level budget alerts so there's one place to look. The 24h delay is exactly why the in-app cap exists: by the time AWS notices the anomaly, the loop already spent the money. Recipes and the alert-routing pattern are in `references/cloud-caps.md`; the cap plumbing in depth is `aws-essentials` / `gcp-essentials`.
+
+## Build vs buy
+
+| You want | Use | Trade-off |
+|---|---|---|
+| Zero code change, fastest setup | Helicone (proxy, ~2-min) | adds a network hop / latency |
+| SDK-level capture + a ready cost table | Langfuse (MIT, ships model+tokenizer cost table) | you wire the SDK, but no proxy hop |
+| Hard spend caps at a gateway, per team or app key | Bifrost (self-hosted, Apache-2.0; budgets and rate limits per virtual key) | one more service to run; caps are per key, not per feature |
+| Full control / custom attribution / typed caps | DIY ledger (this skill) | you own pricing-table freshness and capture coverage |
+
+(firecrawl.dev best-llm-observability-tools; guptadeepak.com top-5-llm-observability-platforms-2026, accessed 2026-06-02; docs.getbifrost.ai features/governance/virtual-keys, accessed 2026-10-19.) Buy the proxy/platform when you want spend *visibility* fast; build the ledger when caps and per-feature attribution must live inside your own logic.
+
+## Anti-patterns
+
+| Anti-pattern | Why it's wrong | Do instead |
+|---|---|---|
+| Pricing literals in business logic | a rate change silently mis-bills everything | versioned table, each row dated + sourced |
+| Billing off the pre-send estimate | estimates ignore output/reasoning/cache; off by most of the bill | price the response `usage` object |
+| No idempotency key on ledger rows | retries double-count spend | `request_id` PRIMARY KEY, upsert not insert |
+| Unknown model defaults to $0 | a new model bills as free; leak is invisible | fail loud on a model absent from the table |
+| Only a soft alert, no hard cap | alert fires, loop keeps spending | a hard cap that refuses with a typed error |
+| Org-total budget, no attribution | "spend is up" — but you can't find the leak | tag every row by user/tenant/feature |
+| Counting input tokens only | output+reasoning are 4-5x the cost — the expensive half | capture all token fields from `usage` |
+| Trusting cloud alerts for real-time control | ~24h delay; the loop already spent it | app-level cap is the guard; cloud is the backstop |
+| "Add caching later" with no measurement | savings unproven; may not even hit | instrument `cache_read_tokens`, compute break-even |
+
+## Verify the artifact
+
+`scripts/verify.sh [path]` lints a candidate cost config/ledger (yaml/json/ts) and fails if: a pricing entry lacks `effective_date` or `source`; a model referenced in logic is missing from the table; the ledger schema lacks an idempotency/request key or any attribution key; the budget declares no soft+hard pair; or cost looks derived from a `len()`/char estimate instead of a `usage` field. It is read-only and exits 0 on a clean config and on no config found — no false failure.

package/skills/shared/cost-tracking/evals/README.md

@@ -0,0 +1,3 @@
+# Evals for cost-tracking
+
+These cases are eyeballed, not automated. Read `cases.yaml` and check three things by hand. First, every `should_trigger` prompt should make this skill the obvious pick — pay attention to the non-obvious ones (the "bill is 3x the estimate / reconcile billed vs response usage" line and the Spanish budget-cap line), since those are where a thinner trigger set would miss. Second, every `should_not_trigger` prompt should route cleanly to the named sibling (finance-ops, unit-economics, bookkeeping, pricing, analytics) — if any of those feels like it could legitimately land here, the description boundary needs tightening. Third, run the `capability` scenario against a generated answer and confirm each `must_include` rubric item is actually present, not hand-waved (especially: cost read from `usage`, dated+sourced pricing table, idempotency + attribution on the ledger, and a soft *and* hard cap). Note that `scripts/verify.sh` checks the emitted artifact (the pricing/ledger/budget config), not the prose — a great write-up that emits no checkable config has not finished the job.

package/skills/shared/cost-tracking/evals/cases.yaml

@@ -0,0 +1,45 @@
+skill: cost-tracking
+
+should_trigger:
+  - prompt: "Wrap our OpenAI calls so we record token cost per request."
+    why: Core capture — per-request token + cost capture around an SDK call is the skill's first step.
+  - prompt: "Set a $500/month cap on the Claude API that hard-stops when we hit it."
+    why: Budget plus a hard cap that refuses — the guard layer this skill owns.
+  - prompt: "Our bill is 3x what we estimated — reconcile the billed usage against what we computed."
+    why: Non-obvious — reconciliation of provider-billed vs ledgered usage is a direct tell for this skill, not generic debugging.
+  - prompt: "Attribute our LLM spend per tenant so we know which customer is expensive."
+    why: Ledger attribution — per-tenant cost rollup is exactly the attribution-key design here.
+  - prompt: "Pon un tope de gasto a la API y avísame al 80%."
+    why: Spanish — budget plus a soft alert threshold (80%), the alert/cap branch of the skill.
+  - prompt: "Is prompt caching actually cheaper for our agent, and where's the break-even?"
+    why: The levers section — caching break-even math is owned here, not asserted.
+  - prompt: "A runaway agent loop burned through our token budget overnight with no guardrail."
+    why: The surprise-bill scenario the hard cap exists to prevent; routes straight to budgets+caps.
+
+should_not_trigger:
+  - prompt: "What's our cash runway at this burn rate?"
+    route_to: finance-ops
+    why: Company-level cash, not the AI/cloud line item; cost-tracking is just one input feeding it.
+  - prompt: "Compute the gross margin per subscription."
+    route_to: unit-economics
+    why: Margin math — cost-tracking only produces the cost numerator, it does not divide by units.
+  - prompt: "Record this AWS invoice as an expense in the books."
+    route_to: bookkeeping
+    why: The accounting record, not the operational real-time meter.
+  - prompt: "Decide what to charge customers for the pro plan."
+    route_to: pricing
+    why: Willingness-to-pay / what to charge, not controlling what we spend.
+  - prompt: "Chart our monthly active users and retention curve."
+    route_to: analytics
+    why: Product/behavior telemetry, a different meter; the dollar capture is here but user-behavior charting is not.
+
+capability:
+  - scenario: "We have a Next.js app calling Claude + OpenAI with no cost visibility and a scary bill. Give us a metering + budget-cap design."
+    must_include:
+      - reads cost from the response `usage` object, not pre-send estimates (estimates only for the pre-flight cap check)
+      - versioned pricing table where each model row carries effective_date + source, loaded as data not literals
+      - loud failure on a model absent from the pricing table (never default to $0)
+      - append-only ledger with an idempotency/request key and at least one attribution key (user/tenant/feature)
+      - budget with BOTH a soft threshold (alert/degrade) and a hard threshold (refuse with a typed error)
+      - the three levers (prompt caching with ~1-hit break-even, batch API 50% stacking with caching, model downshift)
+      - notes cloud billing alerts (AWS CAD ~24h delay, GCP Pub/Sub) are a delayed backstop, not the real-time guard

package/skills/shared/cost-tracking/references/cloud-caps.md

@@ -0,0 +1,52 @@
+# Cloud spend caps — recipes (accessed 2026-06-02)
+
+App-level metering is the real-time guard. Everything here is the **slow backstop** — useful for catching what the app meter missed, never the thing standing between you and a runaway loop. The recurring caveat: cloud billing data lags, so by the time the cloud notices, the money is already spent.
+
+## AWS
+
+### Cost Anomaly Detection (ML-based)
+- ML model, runs roughly **3x/day**, with **up to 24h data delay**.
+- Emits to **SNS** → fan out to a **Lambda** that posts into your alert pipe (or revokes a key / disables a feature flag).
+- Source: aws.amazon.com/aws-cost-management/aws-cost-anomaly-detection; docs.aws.amazon.com/cost-management/.../manage-ad.html. Accessed 2026-06-02.
+
+### Budgets (threshold)
+- Set 50/80/100% threshold alerts on a monthly budget → SNS → same Lambda.
+- Pairs with CAD: Budgets catch the *expected* threshold; CAD catches the *unexpected* spike.
+
+```text
+AWS Budgets / Cost Anomaly Detection
+        │ (SNS, up to ~24h delay)
+        ▼
+     Lambda  ──►  shared alert pipe (Slack/PagerDuty/webhook)
+              └─► optional: disable feature flag / rotate-revoke API key
+```
+
+## GCP
+
+- Budgets are **threshold-only** for alerting, but the budget can publish to **Pub/Sub**.
+- Pub/Sub → **Cloud Function** is the one path that can actually **pause or throttle a workload** programmatically (e.g. scale a service to zero, disable a billing account binding).
+- Source: costimizer.ai gcp guide. Accessed 2026-06-02.
+
+```text
+GCP Budget ──► Pub/Sub ──► Cloud Function
+                            ├─► shared alert pipe
+                            └─► throttle / pause workload (real control)
+```
+
+## Azure
+
+- Threshold-only budgets with action groups. Same shape as GCP minus the clean Pub/Sub→pause path; wire the action group into the shared alert pipe.
+
+## Alert-routing pattern (all providers)
+
+Route **every** cloud alert into the **same pipe** as your in-app budget alerts, so there is one place to look and one severity ladder. Tag each alert with its source (`aws-cad`, `gcp-budget`, `app-meter`) so reconciliation can tell which guard fired.
+
+## The delay caveat, stated plainly
+
+| Guard | Latency | Can it stop spend? |
+|---|---|---|
+| App-level cap (this skill) | real-time | yes — refuse the request now |
+| AWS CAD / Budgets | up to ~24h | no — alert only (Lambda can revoke after the fact) |
+| GCP budget → Pub/Sub → Function | minutes-to-hours | partial — can pause a workload, but on lagged data |
+
+Design accordingly: the cloud cap is insurance against what your app meter failed to count; it is not the meter.

package/skills/shared/cost-tracking/references/pricing-tables.md

@@ -0,0 +1,51 @@
+# Pricing tables — PERISHABLE snapshot (dated 2026-06-02)
+
+These rates move month to month. **Treat this file as data with an expiry, not a constant.** Every number below carries a source; re-fetch before trusting. The whole point of the skill is that prices live in a dated table, never as literals in logic.
+
+## Anthropic (Claude) — per 1M tokens
+
+| Model | Input | Output | Cache read | Source |
+|---|---|---|---|---|
+| Haiku 4.5 | $1.00 | $5.00 | $0.10 | cloudzero.com/blog/claude-api-pricing |
+| Sonnet 4.6 | $3.00 | $15.00 | $0.30 | cloudzero.com/blog/claude-api-pricing |
+| Opus 4.7 | $5.00 | $25.00 | $0.50 | evolink.ai/blog/claude-api-pricing-guide-2026 |
+
+Cache writes: 5-minute = 1.25x base input; 1-hour = 2x base input. Cache reads = 0.1x base input. Break-even on a 5-min cache entry: after ~1 hit (platform.claude.com prompt-caching; finout.io anthropic-api-pricing). All accessed 2026-06-02.
+
+## OpenAI — per 1M tokens
+
+| Model | Input | Cached input | Source |
+|---|---|---|---|
+| GPT-5.5 | $5.00 | $0.50 | openai.com/api/pricing |
+| GPT-5.4 | $2.50 | $0.25 | devtk.ai/en/blog/openai-api-pricing-guide-2026 |
+
+Cached input = **90% off** standard input across the line. Output rates vary by model — read them off the source, do not assume. Model names and exact rates drift fast; this is precisely why the rate table is data and never a literal. Accessed 2026-06-02.
+
+## Bedrock note
+
+AWS Bedrock re-prices the same underlying models (often with its own regional/commitment rates) and bills on its own line. Do not assume Bedrock matches first-party Anthropic/OpenAI rates — pull the Bedrock model rate separately and tag the table row `source: bedrock`. Cloud-side caps for Bedrock spend belong in `cloud-caps.md`.
+
+## Batch API discount
+
+50% off on both OpenAI and Anthropic, returned async within 24h, and **stacks with prompt caching** (combined up to ~95% off). Source: finout.io anthropic-api-pricing, accessed 2026-06-02. Use for non-realtime work (evals, backfills, nightly jobs).
+
+## The `usage` object — field map per provider
+
+Bill against these, not pre-send estimates.
+
+| Concept | Anthropic | OpenAI |
+|---|---|---|
+| Input tokens | `usage.input_tokens` | `usage.prompt_tokens` |
+| Output tokens | `usage.output_tokens` | `usage.completion_tokens` |
+| Cached (read) | `usage.cache_read_input_tokens` | `usage.prompt_tokens_details.cached_tokens` |
+| Cache write | `usage.cache_creation_input_tokens` | n/a (no explicit write line) |
+| Reasoning | included in output | `usage.completion_tokens_details.reasoning_tokens` |
+
+Pre-flight estimate tools (estimates only): Anthropic `client.messages.count_tokens()` returns input tokens, is free, separate rate limit, **not** tiktoken-compatible; OpenAI/most use `tiktoken` locally. Sources: platform.claude.com token-counting; langfuse.com token-and-cost-tracking. Accessed 2026-06-02.
+
+## How to refresh this file
+
+1. Pull each provider's current pricing page (sources above).
+2. Update the rate and set `effective_date` to today on every changed row.
+3. Add any new model name as a new row — never leave a model out, or the lookup will hit the "unknown model" loud-fail path (which is correct behavior).
+4. Re-run `scripts/verify.sh` against your `pricing.yaml` to confirm every row still carries `effective_date` + `source`.

package/skills/shared/cost-tracking/scripts/verify.sh

@@ -0,0 +1,135 @@
+#!/usr/bin/env bash
+#
+# verify.sh - lint a candidate cost-tracking config/ledger artifact.
+#
+# Usage:
+#   cd <dir containing the config>
+#   ./verify.sh [path-to-config]      # yaml | yml | json | ts | js
+#
+# Read-only: never mutates anything. Checks the artifact for:
+#   1. every pricing entry carries effective_date AND source
+#   2. no model referenced in logic is absent from the pricing table
+#   3. the ledger schema has an idempotency/request key + >=1 attribution key
+#   4. the budget declares BOTH a soft and a hard threshold
+#   5. cost is derived from a usage/response field, not a len()/char estimate
+#
+# Emits [ ok ]/[fail] per check, then PASS or FAIL.
+# Exits 0 on a clean config AND on no config found (no false failure).
+# Exits 1 with the missing pieces on any violation.
+#
+# Portable: stock macOS bash 3.2 + grep + awk. No associative arrays, no mapfile.
+
+set -euo pipefail
+
+if [ -t 1 ]; then
+  RED=$'\033[31m'; GREEN=$'\033[32m'; YELLOW=$'\033[33m'; RESET=$'\033[0m'
+else
+  RED=''; GREEN=''; YELLOW=''; RESET=''
+fi
+
+failed=0
+fail() { printf '%s[fail]%s %s\n' "$RED" "$RESET" "$*"; failed=1; }
+ok()   { printf '%s[ ok ]%s %s\n' "$GREEN" "$RESET" "$*"; }
+
+# --- locate the config -------------------------------------------------------
+file="${1:-}"
+if [ -z "$file" ]; then
+  for cand in cost-tracking.yaml cost-tracking.yml pricing.yaml pricing.yml \
+              cost-config.yaml cost-config.yml cost-tracking.json cost-tracking.ts; do
+    if [ -f "$cand" ]; then file="$cand"; break; fi
+  done
+fi
+
+if [ -z "$file" ] || [ ! -f "$file" ]; then
+  printf '%s[skip]%s no cost-tracking config found (cost-tracking.{yaml,json,ts} / pricing.yaml) - nothing to verify.\n' "$YELLOW" "$RESET"
+  exit 0
+fi
+
+printf -- '----- linting %s\n' "$file"
+
+# case-insensitive fixed-ish grep helper
+has() { grep -Eiq "$1" "$file"; }
+
+# --- 1. pricing entries carry effective_date AND source ----------------------
+mentions_pricing=0
+if has 'pric|input_per_mtok|in_per_mtok|rate_in|cost_per'; then mentions_pricing=1; fi
+if [ "$mentions_pricing" -eq 1 ]; then
+  miss=""
+  has 'effective_date|effective-date|effectiveDate' || miss="$miss effective_date"
+  has 'source' || miss="$miss source"
+  if [ -z "$miss" ]; then
+    ok "pricing entries carry effective_date + source"
+  else
+    fail "pricing table missing:$miss (each model row must be dated and sourced)"
+  fi
+else
+  printf '%s[skip]%s no pricing table in this file (check 1)\n' "$YELLOW" "$RESET"
+fi
+
+# --- 2. no model referenced in logic absent from the table -------------------
+# Heuristic: collect model-ish identifiers (claude-*/gpt-*/sonnet/opus/haiku/o-series)
+# that appear OUTSIDE a "model:" / "- model:" declaration line, and warn if the
+# file declares no model rows at all while still naming models.
+declared=$(grep -Eio '(^|[^a-z])model[[:space:]]*[:=]' "$file" | wc -l | tr -d ' ')
+named=$(grep -Eio '(claude-[a-z0-9.]+|gpt-[a-z0-9.]+|sonnet|opus|haiku|gemini-[a-z0-9.]+)' "$file" \
+        | sort -u | wc -l | tr -d ' ')
+if [ "$named" -gt 0 ]; then
+  if [ "$declared" -eq 0 ]; then
+    fail "models are named but no 'model:' table rows declared - unknown models would price as \$0"
+  else
+    ok "models named ($named) are declared in 'model:' rows ($declared) - lookup can fail-loud on unknowns"
+  fi
+else
+  printf '%s[skip]%s no model identifiers found (check 2)\n' "$YELLOW" "$RESET"
+fi
+
+# --- 3. ledger: idempotency key + >=1 attribution key ------------------------
+mentions_ledger=0
+if has 'ledger|usage|input_tokens|output_tokens|cost_usd'; then mentions_ledger=1; fi
+if [ "$mentions_ledger" -eq 1 ]; then
+  miss=""
+  has 'request_id|idempotency|request-id|requestId|dedup' || miss="$miss idempotency/request key"
+  has 'user_id|tenant_id|tenant|feature|user-id|tenantId|userId' || miss="$miss attribution-key(user/tenant/feature)"
+  if [ -z "$miss" ]; then
+    ok "ledger has an idempotency key + at least one attribution key"
+  else
+    fail "ledger missing:$miss"
+  fi
+else
+  printf '%s[skip]%s no ledger schema in this file (check 3)\n' "$YELLOW" "$RESET"
+fi
+
+# --- 4. budget: both a soft and a hard threshold -----------------------------
+mentions_budget=0
+if has 'budget|threshold|cap|alert'; then mentions_budget=1; fi
+if [ "$mentions_budget" -eq 1 ]; then
+  miss=""
+  has 'soft|alert|warn|degrade|downshift' || miss="$miss soft(alert/degrade)"
+  has 'hard|refuse|hard_cap|hard-cap|block|reject' || miss="$miss hard(refuse)"
+  if [ -z "$miss" ]; then
+    ok "budget declares both a soft and a hard threshold"
+  else
+    fail "budget missing:$miss threshold"
+  fi
+else
+  printf '%s[skip]%s no budget config in this file (check 4)\n' "$YELLOW" "$RESET"
+fi
+
+# --- 5. cost derived from usage, not a len()/char estimate -------------------
+if has 'usage|input_tokens|output_tokens|cached_tokens|prompt_tokens'; then
+  if grep -Eiq 'len\(|\.length|char_count|charcount|//[[:space:]]*4|/[[:space:]]*4' "$file" \
+     && ! has 'estimate|pre-?flight|preflight'; then
+    fail "cost may be derived from a len()/char estimate - bill against the response usage object"
+  else
+    ok "cost derived from a usage/response field, not a char estimate"
+  fi
+else
+  printf '%s[skip]%s no usage-based cost field found (check 5)\n' "$YELLOW" "$RESET"
+fi
+
+echo
+if [ "$failed" -ne 0 ]; then
+  printf '%sFAIL:%s cost-tracking config is missing required pieces (see [fail] lines above).\n' "$RED" "$RESET"
+  exit 1
+fi
+printf '%sPASS:%s cost-tracking config carries the required guardrails.\n' "$GREEN" "$RESET"