| name | forge-reliability |
| description | Audit timeouts, retries, overload, dependencies, degradation, consistency, and operational objectives. Use for services and critical workflows with availability or durability expectations. |
forge-reliability: Reliability
Purpose
Audit timeouts, retries, overload, dependencies, degradation, consistency, and operational objectives.
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-reliability, asks about reliability, or
discovery finds an applicable boundary. Run it from the repository root after project discovery.
When it applies
- Services and critical workflows with availability or durability expectations
When it does not apply
- Disposable local-only utilities with no reliability objective
Do not silently skip it. Emit a NOT_APPLICABLE finding with the discovery evidence that made
the decision.
Inputs from project discovery
- dependency graph
- SLOs
- failure handling and incident history
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
- Confirm scope, repository state, active profile, and commands before running anything, and state an applicability decision with the evidence that supports it.
- Enumerate external dependencies and verify every remote call has a timeout, a bounded retry with idempotency, and a defined failure behavior.
- Trace one dependency outage end to end: what the user sees, what degrades, and what data risks inconsistency.
- Inspect health and readiness checks for truthfulness (they verify real dependencies) and safe shutdown for in-flight work.
- Check retry-storm and thundering-herd protections at every retry site, and circuit breaking where cascades are possible.
- Verify stated SLOs, recovery procedures, and consistency guarantees against implementation evidence, not documentation.
- 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.
- 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
- Define user-visible service indicators and error-budget expectations
- Inspect timeout budgets, bounded retries, jitter, circuit behavior, health checks, overload, bulkheads, and dependency fallbacks
- Check graceful shutdown, deploy overlap, consistency, partial failure, idempotency, and clock assumptions
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.
- Timeouts
- Retries
- Retry storms
- Circuit breaking
- Graceful degradation
- Health checks
- Readiness checks
- Dependency failures
- Partial outages
- Failover
- Idempotency
- Load shedding
- Safe shutdown
- Data consistency
- SLOs
- Recovery procedures
- Regional failures
- Feature degradation
- User-visible failure behavior
Safe executable checks
- Run
forge reliability audit --json or fullstack-forge reliability audit --json when
the CLI is installed.
- Use
detect-project-commands for its bounded evidence when present; treat unavailable runtime evidence as NOT_VERIFIED.
- Use
run-project-command 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 known failure modes and incident learnings
- Confirm provider SLAs and operator escalation paths
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-RELI-001, FF-RELI-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 bounded timeouts, retry classification, and graceful shutdown
- Add health checks that reflect readiness without exposing internals
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
- Changing consistency, fail-open behavior, provider topology, or availability targets
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
- Inject dependency, timeout, overload, and shutdown failures
- Confirm recovery without duplicate durable effects
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
- Google SRE principles
- RFC 9110
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
- Align platform health and termination semantics with application lifecycle
Adapt filenames and commands to detected evidence. Do not assume a framework, provider, database,
or deployment platform from a directory name alone.
Known limitations
- Availability claims require measured production evidence
Completion contract
Never declare a feature complete merely because code was written. A task is complete only when:
- The requested behavior is implemented.
- Relevant workflows work end to end.
- Authentication and authorization are verified.
- Database behavior is reviewed.
- Loading, empty, error, and success states exist.
- Applicable accessibility requirements are addressed.
- Automated checks pass.
- Security-sensitive changes receive security review.
- Performance-sensitive changes receive performance review.
- Remaining risks, skipped checks, and assumptions are reported.
Never hide failed checks or claim that an operation ran when it did not.