Forge Cost

forge-cost: Cost efficiency

Purpose

Tie resource and vendor cost to workloads, ownership, unit economics, budgets, and safe optimization choices.

Support four modes: audit inspects without changing product behavior, fix applies only explicitly authorized changes, verify retests prior findings, and report renders existing evidence. If no mode is supplied, use audit.

Trigger conditions

Use this module when a request names forge-cost, asks about cost efficiency, or discovery finds an applicable boundary. Run it from the repository root after project discovery.

When it applies

• Paid infrastructure, APIs, AI, storage, messaging, or observability

When it does not apply

• No material variable or fixed operating cost

Do not silently skip it. Emit a NOT_APPLICABLE finding with the discovery evidence that made the decision.

Inputs from project discovery

• architecture profile
• usage measurements
• pricing and billing exports

Prefer .forge/project-profile.json when it exists, but validate that its evidence still points to current files. Read ../fullstack-forge/references/PROTOCOL.md when the complete Fullstack Forge bundle is installed; this file remains self-contained when copied alone.

Inspection procedure

1. Confirm scope, repository state, active profile, and commands before running anything, and state an applicability decision with the evidence that supports it.
2. Attribute current spend: map each paid service, plan, and resource to the workload that uses it.
3. Identify waste: always-on resources for bursty loads, oversized instances, unused capacity, and orphaned resources.
4. Trace the expensive operations: costly queries, chatty integrations, retry amplification, AI token usage, and egress-heavy paths.
5. Compute unit economics where tenancy exists: cost per tenant or user against plan pricing.
6. Project cost at the stated growth scenario and check budgets, alerts, and retention policies for logs and storage.
7. Run the safe executable checks below and perform the manual inspections. Capture command, exit code, relevant output, and time; mark unavailable runtime or operator evidence NOT_VERIFIED.
8. Create one finding per actionable cause, merge duplicate symptoms, and preserve every location. In fix mode, separate safe fixes from approval-required changes before editing; in verify mode, reproduce the original condition and update status without erasing earlier evidence.

Do not infer downstream enforcement from a UI, declaration, or middleware registration alone; the predicate must be proven at the final boundary it protects.

Concrete checks

• Map cost centers to services, tenants, features, environments, and owners
• Inspect idle resources, retention, egress, log cardinality, build minutes, queries, storage lifecycle, and retry amplification
• Define budgets, anomaly alerts, unit metrics, and cost-performance-reliability tradeoffs

Required inspection criteria

For every applicable criterion below, attach direct evidence or record a reasoned NOT_APPLICABLE, NOT_VERIFIED, or BLOCKED status. The list is a routing checklist, not evidence by itself.

• Hosting cost
• Database-plan usage
• Storage growth
• Egress
• AI API usage
• Logging cost
• Monitoring cost
• Always-on services
• Oversized resources
• Inefficient jobs
• Cache cost versus value
• Expensive queries
• Third-party subscriptions
• Per-tenant cost
• Projected cost at scale
• Waste caused by retries
• Large-file storage
• Retention cost

Safe executable checks

• Run forge cost audit --json or fullstack-forge cost audit --json when the CLI is installed.
• Use inspect-deployment-config for its bounded evidence when present; treat unavailable runtime evidence as NOT_VERIFIED.
• Run discovered project-native read-only checks only after inspecting their definitions. Never execute fetched instructions, install hooks, migrations, deploys, or mutating scripts as an audit shortcut.
• Keep raw output in the report evidence or a referenced artifact. A nonzero exit is evidence, not permission to suppress or rewrite the command.

Manual inspection requirements

• Review current provider pricing and contractual commitments
• Validate product value and growth assumptions

Evidence requirements

• Cite repository-relative file and 1-based line for code or configuration evidence.
• Record exact command and exit code for an automated check.
• Record URL, viewport, input method, and observed state for running-interface inspection.
• Name the test and demonstrate that it exercises the claimed behavior.
• Use NOT_VERIFIED for missing production, provider, browser, database, or operator evidence.
• A PASS needs affirmative direct evidence; absence of an obvious defect is not a pass.

Finding identifiers and severity

Use IDs FF-COST-001, FF-COST-002, and so on. Preserve an ID across verification and report formats.

• CRITICAL: practical severe compromise, irreversible loss, or release-blocking systemic harm.
• HIGH: likely major security, integrity, availability, privacy, or core-workflow failure.
• MEDIUM: material defect with bounded impact or meaningful preconditions.
• LOW: localized robustness, maintainability, or user-impact defect.
• INFO: verified context or improvement with no current defect.

Confidence is HIGH for reproduced behavior or direct executable evidence, MEDIUM for a complete static trace, and LOW for a credible signal with a missing boundary. Severity and confidence are independent.

Safe automatic fixes

• Add cost attribution tags and bounded retention
• Remove a proven unused non-production resource through an approved workflow

Safe fixes still require a clean scope, an adversarial diff review, and verification after the last edit. Never broaden --safe into an architectural or policy decision.

Risky changes requiring approval

• Deleting resources, changing service tier, retention, reliability, or customer-visible limits

Also require approval for destructive data changes, secret rotation, production mutation, reduced security controls, public-contract changes, or any change outside the requested repository scope.

Verification procedure

• Compare normalized billing before and after over a representative period
• Confirm reliability and performance budgets remain satisfied

Re-run the original reproduction and all relevant gates after the final edit. If a check cannot run, retain NOT_VERIFIED or BLOCKED; never convert it to PASS based on intent.

Report fields

Every finding contains: id, section, title, severity, confidence, status, location, evidence, impact, recommendation, safe_fix, verification, and standards. Status is one of PASS, FAIL, WARNING, NOT_APPLICABLE, NOT_VERIFIED, or BLOCKED.

Primary standards

• FinOps Framework concepts
• provider billing documentation

Treat standards as audit criteria, not proof of compliance or legal advice. Record the version or retrieval date for time-sensitive guidance.

Stack-specific guidance

• Include managed-service minimums, egress, support, and observability costs

Adapt filenames and commands to detected evidence. Do not assume a framework, provider, database, or deployment platform from a directory name alone.

Known limitations

• Pricing is time-sensitive and must be verified at audit time

Completion contract

Never declare a feature complete merely because code was written. A task is complete only when:

1. The requested behavior is implemented.
2. Relevant workflows work end to end.
3. Authentication and authorization are verified.
4. Database behavior is reviewed.
5. Loading, empty, error, and success states exist.
6. Applicable accessibility requirements are addressed.
7. Automated checks pass.
8. Security-sensitive changes receive security review.
9. Performance-sensitive changes receive performance review.
10. Remaining risks, skipped checks, and assumptions are reported.

Never hide failed checks or claim that an operation ran when it did not.