launch-readiness-auditor

Launch Readiness Auditor

Audit one launch at one lifecycle read. Preflight evaluates readiness/assets plus planned policy and instrumentation red lines; execution evaluates observed launch-window operation; outcome evaluates post-lag proof. There is no cross-time composite.

When This Must Trigger

• Before a committed launch/announcement when go/no-go evidence is needed.
• During/after the launch window when execution quality must be assessed.
• After the declared lag when actual outcomes and learnings are reviewed.

Quick Start

Run RAMP preflight for launch alpha against the registry stage, canon, claims, rules, and event QA.
Run the outcome profile at day 30; keep it separate from the preflight result.

Skill Contract

Reads: one launch, one lifecycle read, registry/canon/claims state, and profile-specific evidence. Writes: only a permissioned v3 artifact. Done when: the selected profile is complete or its exact Unknowns are reported, with no launch execution or registry mutation.

launch-registry owns stage/date/embargo facts. launch-day-conductor executes the runbook. This auditor only judges the frozen evidence.

Data Sources

Instructions

Runtime Reads

• ../../../references/auditor-runbook.md
• ../../../references/scoring-semantics.md
• ../../../references/ramp-benchmark.md
• ../../../references/runtime-invocation.md
• references/auditor-runtime.md

Runtime and Setup

Read ../../../references/auditor-runbook.md, scoring-semantics.md, ramp-benchmark.md, and the RAMP catalog entry. Standalone installs use bundled immutable references/auditor-runtime.md; never fetch mutable main. Before deterministic calls, follow runtime-invocation.md, resolve AARON_SKILLS_ROOT="${CLAUDE_PLUGIN_ROOT:-$(git rev-parse --show-toplevel 2>/dev/null || true)}", and require the scorer, validator, and typed catalogs. If unavailable, return score_state: NOT_SCORED / score_confidence: not_scored with no gate verdict or persistent artifact.

Declare profile/lifecycle read (preflight|execution|outcome), target launch, launch type, market, access model, observation date, and evidence window.

Profile Procedure

• Preflight: score R1–R10, A1–A10, planned M1, and P1. A public pricing page is required only when the declared access model promises public paid availability.
• Execution: score observed M1–M10; do not substitute planned runbook quality for actual execution.
• Outcome: score P2–P10 after the declared lag; do not backfill forecast targets as actuals.

Every observed state needs source/date/type/confidence. Missing applicable evidence is Unknown; catalog-authorized conditional items may be N/A with reason. Run python3 "$AARON_SKILLS_ROOT/scripts/rubric-score.py" score <run.json> on the selected typed profile.

Verify profile-relevant vetoes: RAMP-R1 stage/access contradiction, RAMP-A1 material claim/disclosure failure, RAMP-M1 planned or observed manipulation/embargo/platform violation, and RAMP-P1 demonstrably broken instrumentation on participating surfaces.

§2 RAMP Worked Examples

• Complete preflight, raw 80, no veto/fail: DONE/SHIP, final 80.
• Complete preflight, raw 76, one verified A1 failure: DONE_WITH_CONCERNS/FIX, final 59.
• Complete preflight, verified R1 and M1 failures: DONE/BLOCK, no final score.
• Outcome read before conversion lag or without own-data actuals: NEEDS_INPUT/UNDECIDED, no score.

§3 RAMP Guardrails

• Stage truth follows promised access/eligibility; a pricing page is not universal evidence of GA.
• A genuine feedback request is not vote solicitation.
• Privacy-limited modeled measurement may be Partial; broken required instrumentation is P1 Fail.
• Launch stacking/capacity is an M10 finding, not an automatic veto.
• Never average preflight, execution, and outcome profiles or compare their scores as the same construct.

§5 RAMP Translation

State lifecycle read with every result. On trace request, qualify RAMP-R1/A1/M1/P1, especially against colliding ROAS IDs.

Report and Verdict

Begin with the auditor-runbook's exact typed conversation header. Never replace status, verdict, or score_state with prose; list each explicitly missing qualified item as ``ID: unknown``` before findings. Use the stable catalog ID verbatim (for example, RAMP-R1); never substitute an evidence-subcheck label or synthesize a suffixed ID such as RAMP-R1d`.

Lead with lifecycle-specific verdict, target/context/date, score or coverage/interval, confidence, profile detail, linked prior reads, critical evidence, Unknowns, and fix/rerun owner. A preflight SHIP authorizes no external launch action by itself; explicit execution approval remains required.

Validation Checkpoints

• One launch and one lifecycle profile are declared.
• Plan, execution, and outcome evidence were not mixed.
• Expected items only are scored; Unknown/N/A semantics are correct.
• Stage/access, policy, claims, and instrumentation vetoes are positively verified.
• No launch, submission, registry, or campaign side effect occurred.

Persistence

Persist only after explicit authorization to memory/audits/launch/YYYY-MM-DD-<topic>.md. Preserve the scorer's orthogonal status and verdict; validate the complete v3 draft with validate-audit-artifact.py against the intended relative path, persist only through one full-content Write, and revalidate the target per the auditor runbook. Edit/shell/MCP mutations of the reserved sink are unsupported. Create separate files for separate lifecycle reads and link them by launch ID rather than overwriting.

Reference Materials

• RAMP benchmark
• Auditor runbook
• Scoring semantics
• Measurement protocol

Next Best Skill

• Stage/commitment facts: launch-registry
• Asset/technical fixes: launch-asset-packager
• Execute approved plan: launch-day-conductor
• Outcome monitoring: launch-monitor