docs(aw): strengthen guidance for coverage analysis, incomplete metadata, and non-technical personas

[Copilot]

Three gaps identified by the Agent Persona Explorer (Information Worker, PM, QA Tester personas) in the agentic workflow guidance surface.

create-agentic-workflow.md — new §2e: coverage-analysis

Adds a compact artifact-first decision rule inline with the existing 2a–2d pattern sections. Instructs agents to download pre-computed CI coverage artifacts before re-running tests, specifying exact gh run list/gh run download commands, required actions: read permission + actions toolset, and a noop condition for unchanged coverage. Defers to test-coverage.md for the full template.

report.md — incomplete metadata fallback

Adds guidance for stakeholder digests when customer-impact labels or severity classifications are absent: report missing fields as unclassified, fall back to the nearest available grouping dimension, quantify the gap inline, and call noop rather than emit an empty report when all items lack required metadata.

triggers.md — non-technical workflow examples

Adds a reference table of five non-engineering scenarios (PM digest, doc staleness, compliance label audit, governance sign-off check, on-demand stakeholder report) with trigger, safe-output, and noop rules — grounding the documentation for personas outside engineering.

[pelikhan]

@copilot refactor this section into trigger.md and provide a short list in this file

[Copilot]

Done in 3da0a9b. The full decision matrix, compact scenario examples, pattern-specific noop rules, and the workflow_run vs deployment_status callout have all been moved to triggers.md (new Trigger Decision Matrix, Compact Scenario Examples, and Pattern-Specific noop Rules sections). Section 2 in create-agentic-workflow.md now contains only the short bullet list plus a pointer to triggers.md.

[github-actions]

🤖 PR Triage — §28486872548

Field	Value
Category	docs
Risk	low
Score	27 / 100
Priority	low
Action	⏸ defer

Score breakdown: Impact 12/50 · Urgency 5/30 · Quality 10/20

Batch: draft-wip (PRs #42595, #42596, #42597)

Rationale: Draft docs update strengthening workflow guidance (+51/-1, 4 files). Low urgency, no reviews yet. Defer until ready for review.

Labels applied: pr-type:docs pr-risk:low pr-priority:low pr-action:defer pr-batch:draft-wip pr-agent:copilot-swe-agent

Generated by 🔧 PR Triage Agent · 77.2 AIC · ⌖ 9.82 AIC · ⊞ 1.6K · ◷

[github-actions]

Hey @app/copilot-swe-agent 👋 — great work surfacing and addressing the three gaps from the Agent Persona Explorer! The reorganization of the trigger decision matrix into triggers.md (where it more naturally lives), the unclassified-first fallback guidance in report.md, and the non-technical scenario reference table in triggers.md all sharpen the guidance surface meaningfully.

One thing to look at:

• No test files changed — this is a docs-only PR, so conventional unit tests don't apply. However, if the project has any automated validation for .github/aw/ guidance files (e.g. markdown linting, schema checks, or snapshot/fixture tests that reference these docs), it's worth confirming those still pass — particularly given the 36-line deletion from create-agentic-workflow.md and the lock-file hash bump in pr-code-quality-reviewer.lock.yml.

If you'd like a hand verifying:

Check whether the repository has any automated validation targeting files under .github/aw/ — for example, markdown linting rules, schema validators, or snapshot/fixture tests that reference create-agentic-workflow.md, report.md, or triggers.md.
If any such checks exist, run them locally and fix any failures introduced by PR #42596 (the deletion of the decision matrix from create-agentic-workflow.md and the additions to triggers.md and report.md).
If no such checks exist, confirm that the lock-file hash update in .github/workflows/pr-code-quality-reviewer.lock.yml reflects only the expected body_hash change and no unintended manifest changes.

Generated by ✅ Contribution Check · 360.2 AIC · ⌖ 15.7 AIC · ⊞ 6.3K · ◷