AgentRC AI-readiness assessment
Run AgentRC readiness, transform the CommandResult<T> JSON into a tailored static HTML report, and summarize maturity, score, weak pillars, and remediation.
When to invoke
- "Assess this repository's AI readiness."
- "Run an AgentRC readiness audit."
- "Score how AI-ready this repo is."
- "Generate the AI readiness dashboard with a policy."
Inputs
Use $ARGUMENTS for optional scan flags. Accept --policy <path-or-pkg> to apply an org-specific policy and --per-area to request per-area output. If $ARGUMENTS is empty, run built-in AgentRC defaults unless agentrc.config.json declares a policies array.
Prerequisites and context
- Node 20+ must be on
PATH; check with node --version when uncertain.
- AgentRC is invoked with
npx -y github:microsoft/agentrc readiness.
- The skill may create or overwrite only
reports/index.html.
- The report must be self-contained and openable with
file://.
- For a primer on policies, suggest the
acreadiness-policy skill.
Procedure
- Confirm Node 20+ is available with
node --version when the environment is unknown.
- Choose policy input in precedence order: user-provided
--policy <source>, agentrc.config.json policies array, then built-in defaults.
- Run the readiness scan from the repository root with structured output:
npx -y github:microsoft/agentrc readiness --json [--policy <source>] [--per-area]
- Treat the
CommandResult<T> JSON envelope from npx github:microsoft/agentrc readiness as the source data for rendering.
- Hand off interpretation and rendering to the
ai-readiness-reporter custom agent, using the bundled report-template.html to produce reports/index.html.
- Tell the user how to open
reports/index.html and summarize maturity level, overall score, top three lowest pillars, and the highest-leverage next action.
Dashboard content requirements
| Area |
Required content |
| Summary |
Maturity level, overall score, grade, and pass-rate versus threshold. |
| Pillars |
All 9 pillars across Repo Health (8) and AI Setup (1). |
| Per-pillar explanation |
What it measures, why it matters for AI, current state, and one specific recommendation. |
| Relevance |
AI relevance badge: High, Medium, or Low. |
| Extras |
Show separately; extras never affect the score. |
| Active Policy |
Include disabled criteria, overridden criteria, disabled/overridden policy markers, and thresholds. |
| Remediation |
Prioritised Remediation Plan grouped as Fix First, Fix Next, and Plan. |
| Raw data |
Embed the raw AgentRC JSON for reuse. |
Gotchas
- AgentRC has a built-in renderer with
--visual / --output report.html, but this skill intentionally uses the custom dashboard for an opinionated report closer to a code review than a metrics dump.
- For CI gating, recommend
agentrc readiness --fail-level <n> with a level from 1 to 5.
- The almost-always highest-leverage next action is to run the
acreadiness-generate-instructions skill after reviewing the assessment.
Progressive disclosure and bundled resources
report-template.html: static dashboard template used by the ai-readiness-reporter custom agent. Inline all CSS so the report works under file://.
Output template
## AI-readiness assessment result
**Status:** report created | blocked
**Report:** `reports/index.html`
**Command:** `npx -y github:microsoft/agentrc readiness --json <policy/per-area flags>`
| Metric | Value |
| --- | --- |
| Maturity level | `<level>` |
| Overall score | `<score>` |
| Grade | `<grade>` |
| Pass-rate vs threshold | `<value>` |
### Lowest pillars
1. `<pillar>` - <reason>
2. `<pillar>` - <reason>
3. `<pillar>` - <reason>
### Highest-leverage next action
<next action, usually run the acreadiness-generate-instructions skill>
Quality gate
1---2name: acreadiness-assess-23description: Run the AgentRC AI-readiness assessment for the current repository, optionally apply a policy, and produce a self-contained HTML dashboard at reports/index.html. Use when asked to assess, audit, score, or report how AI-ready a repo is.4---56# AgentRC AI-readiness assessment78Run AgentRC readiness, transform the `CommandResult<T>` JSON into a tailored static HTML report, and summarize maturity, score, weak pillars, and remediation.910## When to invoke1112- "Assess this repository's AI readiness."13- "Run an AgentRC readiness audit."14- "Score how AI-ready this repo is."15- "Generate the AI readiness dashboard with a policy."1617## Inputs1819Use `$ARGUMENTS` for optional scan flags. Accept `--policy <path-or-pkg>` to apply an org-specific policy and `--per-area` to request per-area output. If `$ARGUMENTS` is empty, run built-in AgentRC defaults unless `agentrc.config.json` declares a `policies` array.2021## Prerequisites and context2223- Node 20+ must be on `PATH`; check with `node --version` when uncertain.24- AgentRC is invoked with `npx -y github:microsoft/agentrc readiness`.25- The skill may create or overwrite only `reports/index.html`.26- The report must be self-contained and openable with `file://`.27- For a primer on policies, suggest the `acreadiness-policy` skill.2829## Procedure30311. Confirm Node 20+ is available with `node --version` when the environment is unknown.322. Choose policy input in precedence order: user-provided `--policy <source>`, `agentrc.config.json` `policies` array, then built-in defaults.333. Run the readiness scan from the repository root with structured output:3435```bash36npx -y github:microsoft/agentrc readiness --json [--policy <source>] [--per-area]37```38394. Treat the `CommandResult<T>` JSON envelope from `npx github:microsoft/agentrc readiness` as the source data for rendering.405. Hand off interpretation and rendering to the `ai-readiness-reporter` custom agent, using the bundled `report-template.html` to produce `reports/index.html`.416. Tell the user how to open `reports/index.html` and summarize maturity level, overall score, top three lowest pillars, and the highest-leverage next action.4243## Dashboard content requirements4445| Area | Required content |46| --- | --- |47| Summary | Maturity level, overall score, grade, and pass-rate versus threshold. |48| Pillars | All 9 pillars across Repo Health (8) and AI Setup (1). |49| Per-pillar explanation | What it measures, why it matters for AI, current state, and one specific recommendation. |50| Relevance | AI relevance badge: High, Medium, or Low. |51| Extras | Show separately; extras never affect the score. |52| Active Policy | Include disabled criteria, overridden criteria, disabled/overridden policy markers, and thresholds. |53| Remediation | Prioritised Remediation Plan grouped as Fix First, Fix Next, and Plan. |54| Raw data | Embed the raw AgentRC JSON for reuse. |5556## Gotchas5758- AgentRC has a built-in renderer with `--visual` / `--output report.html`, but this skill intentionally uses the custom dashboard for an opinionated report closer to a code review than a metrics dump.59- For CI gating, recommend `agentrc readiness --fail-level <n>` with a level from 1 to 5.60- The almost-always highest-leverage next action is to run the `acreadiness-generate-instructions` skill after reviewing the assessment.6162## Progressive disclosure and bundled resources6364- `report-template.html`: static dashboard template used by the `ai-readiness-reporter` custom agent. Inline all CSS so the report works under `file://`.6566## Output template6768```markdown69## AI-readiness assessment result7071**Status:** report created | blocked72**Report:** `reports/index.html`73**Command:** `npx -y github:microsoft/agentrc readiness --json <policy/per-area flags>`7475| Metric | Value |76| --- | --- |77| Maturity level | `<level>` |78| Overall score | `<score>` |79| Grade | `<grade>` |80| Pass-rate vs threshold | `<value>` |8182### Lowest pillars831. `<pillar>` - <reason>842. `<pillar>` - <reason>853. `<pillar>` - <reason>8687### Highest-leverage next action88<next action, usually run the acreadiness-generate-instructions skill>89```9091## Quality gate9293- [ ] Node 20+ was verified or the environment already proved it.94- [ ] Policy precedence was applied: `$ARGUMENTS`, `agentrc.config.json`, then built-in defaults.95- [ ] Readiness ran with `--json` and preserved the `CommandResult<T>` envelope.96- [ ] `reports/index.html` was produced from `report-template.html` and is self-contained.97- [ ] The report includes all 9 pillars, extras, Active Policy details, remediation priorities, and raw JSON.98- [ ] No repository file other than `reports/index.html` was modified.
paulasilvatech