feat: update workflows to AI credits billing terminology

Author: mnkiefer

.github/workflows/agentic-token-audit.md

@@ -1,5 +1,5 @@
 ---
-description: Daily audit of token usage across all agentic workflows with historical trend tracking
+description: Daily audit of AI credit spend across all agentic workflows with historical trend tracking
 on:
   schedule:
     - cron: "daily around 12:00 on weekdays"
@@ -71,15 +71,15 @@ steps:
 timeout-minutes: 25
 ---
 
-# Daily Agentic Workflow Token Usage Audit
+# Daily Agentic Workflow AI Credit Spend Audit
 
-You are the Agentic Workflow Token Auditor — a workflow that tracks daily token consumption across all agentic workflows in this repository and maintains a historical record for trend analysis.
+You are the Agentic Workflow Auditor — a workflow that tracks daily AI credit spend and token consumption across all agentic workflows in this repository and maintains a historical record for trend analysis.
 
 ## Mission
 
-1. Parse the pre-downloaded agentic workflow logs and compute per-workflow token usage metrics.
+1. Parse the pre-downloaded agentic workflow logs and compute per-workflow AI credit spend and token usage metrics.
 2. Persist today's snapshot to repo-memory so the optimizer (and future runs of this audit) can read historical data.
-3. Publish a concise audit issue summarizing today's usage and trend highlights.
+3. Publish a concise audit issue summarizing today's AI credit spend and trend highlights.
 
 ## Data Sources
 
@@ -103,8 +103,9 @@ Each element of `.runs` is a `RunData` object with (among others):
 |---|---|---|
 | `workflow_name` | string | Human-readable name |
 | `workflow_path` | string | `.github/workflows/....lock.yml` |
+| `ai_credits` | float | AI credits consumed (primary billing metric; 1 AIC = $0.01 USD) |
 | `token_usage` | int | Total tokens (`omitempty` — treat missing/null as 0) |
-| `effective_tokens` | int | Normalized token metric |
+| `effective_tokens` | int | Legacy normalized token metric (deprecated; use `ai_credits` for billing) |
 | `action_minutes` | float | Billable GitHub Actions minutes |
 | `turns` | int | Number of agent turns |
 | `duration` | string | Human-readable duration |
@@ -128,9 +129,9 @@ Write a Python script to `/tmp/gh-aw/token-audit/process_audit.py` and run it. T
 1. Load `/tmp/gh-aw/token-audit/workflow-logs.json` and extract `.runs`.
 2. Filter to `status == "completed"` runs only.
 3. Group by `workflow_name` and compute per-workflow aggregates:
-   - `run_count`, `total_tokens`, `avg_tokens`, `total_turns`, `avg_turns`, `total_action_minutes`, `error_count`, `warning_count`
-4. Compute an overall summary: total runs, total tokens, total action minutes.
-5. Sort workflows descending by `total_tokens`.
+   - `run_count`, `total_ai_credits`, `avg_ai_credits`, `total_tokens`, `avg_tokens`, `total_turns`, `avg_turns`, `total_action_minutes`, `error_count`, `warning_count`
+4. Compute an overall summary: total runs, total AI credits, total tokens, total action minutes.
+5. Sort workflows descending by `total_ai_credits`.
 6. Save the result to `/tmp/gh-aw/token-audit/audit_snapshot.json` with this shape:
 
 ```json
@@ -139,13 +140,16 @@ Write a Python script to `/tmp/gh-aw/token-audit/process_audit.py` and run it. T
   "period_days": 30,
   "overall": {
     "total_runs": N,
+    "total_ai_credits": F,
     "total_tokens": N,
     "total_action_minutes": F
   },
   "workflows": [
     {
       "workflow_name": "...",
       "run_count": N,
+      "total_ai_credits": F,
+      "avg_ai_credits": F,
       "total_tokens": N,
       "avg_tokens": N,
       "total_turns": N,
@@ -159,15 +163,15 @@ Write a Python script to `/tmp/gh-aw/token-audit/process_audit.py` and run it. T
 }
 ```
 
-Handle null/missing `token_usage` by treating them as 0.
+Handle null/missing `ai_credits` and `token_usage` by treating them as 0.
 
 ## Phase 2 — Persist Snapshot to Repo-Memory
 
 1. Read the snapshot from `/tmp/gh-aw/token-audit/audit_snapshot.json`.
 2. Copy it to `/tmp/gh-aw/repo-memory/default/YYYY-MM-DD.json` (today's UTC date).
 3. This file is what the optimizer workflow reads to identify high-usage workflows.
 
-Also maintain a rolling summary file at `/tmp/gh-aw/repo-memory/default/rolling-summary.json` that contains an array of daily overall totals (date, total_tokens, total_runs, total_action_minutes) for the last 90 entries. Load the existing file, append today's entry, trim to 90, and save.
+Also maintain a rolling summary file at `/tmp/gh-aw/repo-memory/default/rolling-summary.json` that contains an array of daily overall totals (date, total_ai_credits, total_tokens, total_runs, total_action_minutes) for the last 90 entries. Load the existing file, append today's entry, trim to 90, and save.
 
 Do not append a synthetic zero-valued entry to `rolling-summary.json` when either of these conditions is true:
 
@@ -180,8 +184,8 @@ Report those two cases differently in the issue as described below so the empty-
 
 Create up to two chart images in `/tmp/gh-aw/token-audit/charts/` using Python, `matplotlib`, and `seaborn` with `whitegrid` styling:
 
-1. **Token usage by workflow** (`token_by_workflow.png`): a horizontal bar chart of the top 15 workflows by total tokens from `audit_snapshot.json`.
-2. **Historical token trend** (`token_trend.png`): a line chart from `rolling-summary.json`.
+1. **AI credit spend by workflow** (`ai_credits_by_workflow.png`): a horizontal bar chart of the top 15 workflows by total AI credits from `audit_snapshot.json`.
+2. **Historical AI credit trend** (`ai_credits_trend.png`): a line chart from `rolling-summary.json`.
 
 Chart requirements:
 
@@ -191,8 +195,8 @@ Chart requirements:
 - Save only PNG files.
 - If there are fewer than 2 rolling-summary points, skip the trend chart and explain why in the issue.
 - After generating each chart, call `upload_asset` with its file path.
-- In the issue template below, replace `UPLOAD_URL_WORKFLOW_PLACEHOLDER` with the URL returned for `token_by_workflow.png`.
-- In the issue template below, replace `UPLOAD_URL_TREND_PLACEHOLDER` with the URL returned for `token_trend.png`.
+- In the issue template below, replace `UPLOAD_URL_WORKFLOW_PLACEHOLDER` with the URL returned for `ai_credits_by_workflow.png`.
+- In the issue template below, replace `UPLOAD_URL_TREND_PLACEHOLDER` with the URL returned for `ai_credits_trend.png`.
 - If a chart is skipped, omit that image markdown line entirely instead of leaving a placeholder behind.
 
 ## Phase 4 — Publish Audit Issue
@@ -213,45 +217,46 @@ Create an issue with these sections:
 
 - **Period**: last 24 hours (YYYY-MM-DD to YYYY-MM-DD)
 - **Total runs**: N
+- **Total AI credits**: N.NN AIC
 - **Total tokens**: N (formatted with commas)
 - **Total Actions minutes**: X.X min
 - **Active workflows**: N
 
-### 🏆 Top 5 Workflows by Token Usage
+### 🏆 Top 5 Workflows by AI Credit Spend
 
-| Workflow | Runs | Total Tokens | Avg Tokens |
+| Workflow | Runs | Total AI Credits | Avg AI Credits |
 |---|---|---|---|
 | ... | ... | ... | ... |
 
 ### 📈 Trends
 
 Embed chart images using uploaded asset URLs when available:
 
-![Token Usage by Workflow](UPLOAD_URL_WORKFLOW_PLACEHOLDER)
+![AI Credit Spend by Workflow](UPLOAD_URL_WORKFLOW_PLACEHOLDER)
 
-![Historical Token Trend](UPLOAD_URL_TREND_PLACEHOLDER)
+![Historical AI Credit Trend](UPLOAD_URL_TREND_PLACEHOLDER)
 
-Summarize token changes from `rolling-summary.json` when historical data is available.
+Summarize AI credit and token changes from `rolling-summary.json` when historical data is available.
 
 <details>
 <summary><b>Full Per-Workflow Breakdown</b></summary>
 
-[Complete table of all workflows sorted by total tokens]
+[Complete table of all workflows sorted by total AI credits]
 
 </details>
 
 ### 💡 Observations
 
-- Identify any workflow with >30% of total tokens as a "heavy hitter"
+- Identify any workflow with >30% of total AI credits as a "heavy hitter"
 - Note workflows with high error/warning counts relative to runs
-- Flag any workflow whose avg tokens per run exceeds 100,000
+- Flag any workflow whose avg AI credits per run exceeds 1.00 AIC
 
 **Data snapshot**: `memory/token-audit/YYYY-MM-DD.json`
 ```
 
 ## Important Notes
 
-- Use `// 0` (null coalescing) in jq and `.get(field, 0)` in Python for nullable numeric fields.
+- Use `// 0` (null coalescing) in jq and `.get(field, 0)` in Python for nullable numeric fields (`ai_credits`, `token_usage`).
 - Distinguish between these two cases in the issue:
   - the raw `.runs` array is empty
   - the raw `.runs` array is non-empty but none of the runs are `status == "completed"`

.github/workflows/agentic-token-optimizer.md

@@ -1,5 +1,5 @@
 ---
-description: Daily optimizer that identifies a high-token-usage agentic workflow, audits its runs, and recommends efficiency improvements including inline sub-agent refactors when warranted
+description: Daily optimizer that identifies high-AI-credit-spend agentic workflows, audits their runs, and recommends efficiency improvements including inline sub-agent refactors when warranted
 on:
   schedule:
     - cron: "daily around 14:00 on weekdays"
@@ -71,6 +71,7 @@ steps:
             | select(.status == "completed")
             | {
                 workflow_name: .workflow_name,
+                ai_credits: (.ai_credits // 0),
                 tokens: (.token_usage // 0),
                 turns: (.turns // 0),
                 action_minutes: (.action_minutes // 0)
@@ -80,12 +81,14 @@ steps:
           | map({
               workflow_name: .[0].workflow_name,
               run_count: length,
+              total_ai_credits: (map(.ai_credits) | add),
+              avg_ai_credits: ((map(.ai_credits) | add) / length),
               total_tokens: (map(.tokens) | add),
               avg_tokens: ((map(.tokens) | add) / length),
               total_turns: (map(.turns) | add),
               total_action_minutes: (map(.action_minutes) | add)
             })
-          | sort_by(.total_tokens)
+          | sort_by(.total_ai_credits)
           | reverse
           | .[:10]
         )
@@ -107,9 +110,9 @@ steps:
       fi
 ---
 
-# Agentic Workflow Token Usage Optimizer
+# Agentic Workflow AI Credit Spend Optimizer
 
-You are the Agentic Workflow Token Optimizer. Pick one high-token workflow, audit recent runs, and create a conservative optimization issue with measurable improvements. Your recommendations may include prompt, tool, reliability, setup-prefix, and inline sub-agent improvements when the evidence supports them.
+You are the Agentic Workflow Optimizer. Pick one high-AI-credit-spend workflow, audit recent runs, and create a conservative optimization issue with measurable improvements. Your recommendations may include prompt, tool, reliability, setup-prefix, and inline sub-agent improvements when the evidence supports them.
 
 ## Objectives
 
@@ -148,23 +151,24 @@ Prefer `--jq` on `gh api` calls over a separate `| jq` step when the filter is s
 ## Data Inputs
 
 - `/tmp/gh-aw/token-audit/all-runs.json`: full 7-day run data (`gh aw logs --json`).
-- `/tmp/gh-aw/token-audit/top-workflows.json`: pre-aggregated top 10 workflows by total tokens.
+- `/tmp/gh-aw/token-audit/top-workflows.json`: pre-aggregated top 10 workflows by total AI credits.
 - `/tmp/gh-aw/repo-memory/default/YYYY-MM-DD.json`: daily audit snapshots.
 - `/tmp/gh-aw/repo-memory/default/optimization-log.json`: prior optimizations (if present).
 
-Treat missing numeric fields (`token_usage`, `turns`, `action_minutes`) as `0`.
+Treat missing numeric fields (`ai_credits`, `token_usage`, `turns`, `action_minutes`) as `0`.
 
 ## Phase 1 — Select Target
 
 - Start from `top-workflows.json`.
 - Exclude workflows optimized in the last 14 days (use `optimization-log.json`).
 - Exclude workflows with "Token" in the name to avoid self-targeting.
-- Choose the highest token workflow that remains.
+- Choose the highest AI-credit-spend workflow that remains.
 - If no snapshot/history exists, derive candidates directly from `all-runs.json`.
 
 Then collect run-level data for the selected workflow:
 
 - run count
+- total and average AI credits
 - total and average tokens
 - total and average turns
 - conclusions/error patterns
@@ -176,8 +180,8 @@ Use this compact analysis matrix:
 | Area | Required checks | Output |
 |---|---|---|
 | Tool usage | Compare configured tools from workflow source vs observed usage across multiple runs | Keep / Consider removing / Remove |
-| Token efficiency | Evaluate token totals, effective tokens, cache efficiency, turns | Top token waste drivers |
-| Reliability | Repeated errors, warnings, retries, missing tools | Token waste from failures |
+| AI credit spend | Evaluate AI credits, token totals, cache efficiency, turns | Top spend drivers |
+| Reliability | Repeated errors, warnings, retries, missing tools | AI credit waste from failures |
 | Prompt efficiency | Redundant instructions, overlong sections, avoidable iteration | Prompt reduction opportunities |
 | Structural optimization | Repeated setup/tool-call prefixes and sections suited for inline sub-agents | Extract setup / Add sub-agent / Keep in main agent |
 
@@ -279,10 +283,10 @@ Create one issue with:
 
 - **Target workflow + reason selected**
 - **Analysis period + runs analyzed**
-- **Token profile table** (total tokens, avg tokens/run, avg turns/run, cache efficiency)
+- **Spend profile table** (total AI credits, avg AI credits/run, total tokens, avg turns/run, cache efficiency)
 - **Ranked recommendations** with:
   - title
-  - estimated token savings per run
+  - estimated AI credit savings per run
   - concrete action
   - evidence from observed runs
 - **Optional structural optimizations** for shared setup prefixes and inline sub-agents when supported by the analysis
@@ -300,7 +304,7 @@ Create one issue with:
 
 Append one entry to `/tmp/gh-aw/repo-memory/default/optimization-log.json`:
 
-`{"date":"YYYY-MM-DD","workflow_name":"...","total_tokens_analyzed":N,"runs_audited":N,"recommendations_count":N,"subagent_candidates":N,"estimated_savings_per_run":N}`
+`{"date":"YYYY-MM-DD","workflow_name":"...","total_ai_credits_analyzed":F,"total_tokens_analyzed":N,"runs_audited":N,"recommendations_count":N,"subagent_candidates":N,"estimated_ai_credit_savings_per_run":F}`
 
 Use `subagent_candidates` for the count of inline sub-agent candidates you actually recommend in the issue body.
 

CHANGELOG.md

@@ -23,6 +23,12 @@ Changelog entries should describe what changed for users of these workflows, not
 - Omit vague filler like "various fixes" when the actual effect can be named.
 - Do not include contributor handles, reviewer names, or other attribution in release entries.
 
+## [0.3.0] - 2026-06-08
+
+### Changed
+
+- Updated both workflows to track and report AI credits (AIC) as the primary billing metric, aligned with the GitHub Copilot AI credits billing model introduced on June 1, 2026 (1 AIC = $0.01 USD). The daily audit snapshot and rolling summary now include `total_ai_credits`; audit and optimizer issues now lead with AI credit spend. Effective tokens remain available as a legacy compatibility field in the run data but are no longer the primary reporting metric.
+
 ## [0.2.2] - 2026-06-01
 
 ### Changed

README.md

@@ -1,16 +1,16 @@
 # `agentic-ops`
-> "Audit token spend, surface waste, and optimize agentic workflows with confidence!"
+> "Audit AI credit spend, surface waste, and optimize agentic workflows with confidence!"
 
-`agentic-ops` is a focused bundle of GitHub Agentic Workflows for teams scaling agentic automation and wanting better visibility into usage, trends, and optimization opportunities. Instead of guessing which workflows are token-heavy or where token waste is hiding, this package gives you an audit trail, historical reporting, and conservative recommendations you can review before making changes.
+`agentic-ops` is a focused bundle of GitHub Agentic Workflows for teams scaling agentic automation and wanting better visibility into usage, trends, and optimization opportunities. Instead of guessing which workflows are driving the highest AI credit spend or where waste is hiding, this package gives you an audit trail, historical reporting, and conservative recommendations you can review before making changes.
 
 ## Introduction
 
 It is built for platform engineers, developer productivity teams, and repository maintainers who are scaling agentic workflows and need a practical way to keep them efficient. The bundle helps solve a common problem with AI automation: token usage grows quickly, but the signals for where to improve are scattered across workflow runs and logs. With `agentic-ops`, you get repeatable workflows that make usage measurable, optimization opportunities actionable, and efficiency work easier to operationalize.
 
 ## Key Features
 
-- **Clear operational visibility** with a daily token audit that captures usage, trends, and workflow-level hotspots.
-- **Actionable optimization guidance** that identifies high-token workflows and proposes safe, conservative improvements.
+- **Clear operational visibility** with a daily audit that captures AI credit spend, usage trends, and workflow-level hotspots.
+- **Actionable optimization guidance** that identifies high-AI-credit-spend workflows and proposes safe, conservative improvements.
 - **Faster efficiency improvements** by helping teams find waste before it becomes recurring operational overhead.
 - **Built for real GitHub workflows** using GitHub Agentic Workflows, so installation and adoption fit naturally into existing repositories.
 - **Useful historical context** through shared snapshots that support trend analysis instead of one-off debugging.
@@ -41,17 +41,17 @@ Required configuration after installation:
 
 After installation, you can use the included workflows to:
 
-- run a daily audit of workflow token usage
-- identify the workflows consuming the most tokens
+- run a daily audit of AI credit spend and workflow token usage
+- identify the workflows consuming the most AI credits
 - generate optimization recommendations grounded in recent run data
 - surface workflows that are good candidates for inline sub-agent refactors
 
 Included workflows:
 
 | Workflow | What it does |
 | ----- | --- |
-| [`Daily Agentic Workflow Token Usage Audit`](https://github.com/githubnext/agentic-ops/blob/main/workflows/agentic-token-audit.md?plain=1) | Collects recent agentic workflow usage and creates a daily audit snapshot. |
-| [`Agentic Workflow Token Usage Optimizer`](https://github.com/githubnext/agentic-ops/blob/main/workflows/agentic-token-optimizer.md?plain=1) | Analyzes token-heavy workflows and proposes conservative token-reduction changes, including inline sub-agent opportunities when they are a strong fit. |
+| [`Daily Agentic Workflow AI Credit Spend Audit`](https://github.com/githubnext/agentic-ops/blob/main/workflows/agentic-token-audit.md?plain=1) | Collects recent agentic workflow usage and creates a daily AI credit spend snapshot. |
+| [`Agentic Workflow AI Credit Spend Optimizer`](https://github.com/githubnext/agentic-ops/blob/main/workflows/agentic-token-optimizer.md?plain=1) | Analyzes high-AI-credit-spend workflows and proposes conservative efficiency changes, including inline sub-agent opportunities when they are a strong fit. |
 
 ## License
 

aw.yml

@@ -1,6 +1,6 @@
 manifest-version: "1"
 name: agentic-ops
-description: Reusable agentic workflows for auditing and optimizing workflow token usage.
+description: Reusable agentic workflows for auditing AI credit spend and optimizing workflow efficiency.
 emoji: "🤖"
 files:
   - workflows/agentic-token-audit.md