For AI agents: the complete documentation index is available at https://scriptedalchemy.github.io/agent-bundle/llms.txt, the full documentation bundle is available at https://scriptedalchemy.github.io/agent-bundle/llms-full.txt, and this page is available as Markdown at https://scriptedalchemy.github.io/agent-bundle/examples/skills-starter.md.
Skip to main content
agent-bundle
  • English

      • Skills Starter

      A static plugin of Skills, resources, one command, and one rule. It proves that useful host content needs no MCP server, App route, state owner, or executable renderer.

      • Run from the repository root: pnpm example:skills
      • Package: @agent-bundle-example/skills-starter
      • Public dependencies: agent-bundle (workspace:*)
      • Targets: portable, codex, cursor
      • Credentials: none: both eval suites are deterministic and read checked-in fixtures
      • Source: examples/skills-starter

      What it proves

      • Convention discovery. agent-bundle.config.ts declares the plugin and its three targets. Skills, commands, and rules under src/ are found by convention. That is the authoring model described in Skills.
      • One source, three host projections. The same authored documents are lowered for the portable, Codex, and Cursor projections of one composite plugin root, skills/ is emitted once and every selected host reads it. The Workbench's Source and Generated views show whether a host copied a document or adapted it.
      • No invented runtime. The built artifact has no MCP executable, App runtime, Flight worker, state owner, or notice kernel. The Workbench opens static leaves directly and reports unsupported host projections instead of faking an invocation.
      • Deterministic eval evidence. The eval suites run through the deterministic harness against checked-in fixtures, so a Skill's coverage is a recorded run rather than an assertion. Coverage is labeled indirect because a deterministic harness cannot observe host Skill activation.

      What is authored

      PathWhat it is
      src/skills/incident-triage/A production incident from first signal through containment, evidence collection, and a handoff-ready update.
      src/skills/dependency-upgrade/Dependency upgrade planning, implementation, or review with API, runtime, rollout, and rollback checks.
      src/skills/release-review/The evidence, severity, workflow, and final-report requirements for an explicit release review.
      src/commands/review-release.mdA Cursor command that starts the release review.
      src/rules/release-safety.mdcA Cursor rule that keeps release decisions evidence-based.
      evals/release-readiness.eval.tsThe release-readiness suite and its deterministic release-artifact-is-ready case.
      evals/engineering-operations.eval.tsThe engineering-operations suite: incident-handoff-is-actionable and upgrade-plan-has-rollback.

      Each Skill links its own references/ checklist or runbook and a reusable assets/ handoff or planning template, so the Skill body stays short and the long-form material is loaded only when it is needed.

      The dependency-upgrade Skill distinguishes planning, implementation, and release readiness. It selects verification for the affected surfaces instead of requiring every build or consumer check for a plan. Release readiness still needs current candidate evidence and the documented rollback and consumer checks; missing evidence is reported, not treated as success.

      Working in the Workbench

      1. Application → Skills lists dependency-upgrade, incident-triage, and release-review. Select one to render its document, then use the inspector to browse linked checklists and report templates or compare Source and Generated output.
      2. Application → Rules / Commands opens authored content and each host projection directly. Unsupported hosts show their capability reason; there is no Run action or plugin process.
      3. Advanced → Artifact shows the composite plugin root and per-file provenance.
      4. Advanced → Evals → Runs defaults to the release-readiness suite. Run release-artifact-is-ready and inspect the passing trial; it consumes only the checked-in evidence fixture.
      5. To practice repair, make a reversible edit to the release policy, press Rebuild, and wait for a Failed or Idle result rather than a Building state. On failure, open Problems from the header count while Application keeps the last-good tree. Restore the checked-in policy and rebuild. The earlier eval is now stale for the changed build, rerun release-readiness to record current, repaired evidence.

      Noninteractive checks

      After the repository-level pnpm build:

      cd examples/skills-starter
pnpm validate
pnpm build
pnpm exec agent-bundle eval --case release-artifact-is-ready --trials 1

      pnpm check is the validation-and-build pair without the Workbench. Generated output lands in artifact/, and its root contract is artifact/agent-bundle.manifest.json. The .agent-bundle/ directory holds development state and is not source material.