Nw Roadmap Review ChecksSAFE
AI agents that guide you from idea to working code, with you in control at every step.
Overview
AI agents that guide you from idea to working code, with you in control at every step.
1a864ec81cdeOBSERVED · 2026-10-14What it tells the agent
The instruction file, verbatim from the audited commit — this is the text the model reads, and the surface the audit's instruction layer examines. Quoted here so you can judge it without cloning anything.
---
name: nw-roadmap-review-checks
description: Roadmap-specific validation checks for architecture reviews. Load when reviewing roadmaps for implementation readiness.
user-invocable: false
disable-model-invocation: true
---
# Roadmap Review Checks
Six mandatory checks for every roadmap review. Each produces a finding block.
## Check 1: External Validity
Verify completing all steps produces an INVOCABLE feature, not just existing code.
Criteria: at least one step targets entry point integration | acceptance tests invoke through driving ports | clear user invocation path after completion
Severity: BLOCKER if no integration step or invocation path. HIGH if tests at wrong boundary.
```
EXTERNAL VALIDITY CHECK: PASSED|FAILED
Issue: {missing wiring}
Consequence: {what happens without fix}
Required Action: {integration step to add}
```
## Check 2: AC Implementation Coupling
Scan AC for implementation details locking crafter into predetermined structure.
Detection: underscore-prefixed identifiers | method signatures with params | internal class references | specific return types
Severity: HIGH. Flag each coupled AC with rewrite.
- Coupled: `_install_des_module() copies src/des/`
- Rewritten: `DES module importable from installation target after install`
Rationale: AC = WHAT (observable behavior), never HOW (internal structure). Crafter decides during GREEN + REFACTOR.
## Check 3: Step Decomposition Ratio
Calculation: implementation steps / unique production files in `files_to_modify`
Thresholds: acceptable <= 2.0 | warning 2.0-2.5 | reject > 2.5
Identical pattern check: 3+ steps with identical AC differing by substitution -> require batching.
## Check 4: Implementation Code in Roadmap
Verify no implementation code. BLOCKER.
Detection: code snippets/algorithms in descriptions | method implementations in AC | pseudocode/logic | variable names, loops, conditionals
Architect defines WHAT; crafter decides HOW. Roadmap code prevents better solutions via TDD.
## Check 5: Roadmap Concision and Precision
Quantitative thresholds: total words 500 (1-3 steps) / 1500 (4-8) / 3000 (9-15) | step description max 50 words | AC max 5 per step, 30 words each | notes max 100 words
Verbosity detection: multi-sentence blocks -> bullets | qualifiers (comprehensive, robust) -> delete | motivational language -> delete | redundancy -> state once | tutorials -> delete, assume expertise
Precision: every AC single interpretation | testable outcome | business over technical jargon
Severity: BLOCKER if word count exceeds threshold.
## Check 6: Unit Test Boundary Validation
Verify unit tests respect hexagonal boundaries.
Rules: invoke through driving ports only | no domain entity unit tests | behavior focus | never import internal/private modules
- Wrong: `Tests verify TemplateValidator._validate_schema() handles edge cases`
- Right: `When invalid template provided to render_prompt(), ValidationError raised with clear message`
Severity: HIGH.Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | NA |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- none-observed
- Network
- none-observed
- Shell
- none-observed
- Dependencies
- pinned
- Secrets in source
- none-found
Findings (0)
No findings outside the package's declared scope.
Gates applied: no_behavioural_pass.
1a864ec81cdefull audit observations/trust-audit/skill/nwave-ai__nw-roadmap-review-checks.json · Report an issue / request a re-scanAudit history
Every audit this skill has had.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-14 | 1a864ec81cde | SAFE | B | 89 | first audit |
Questions
What does the Nw Roadmap Review Checks skill do?
AI agents that guide you from idea to working code, with you in control at every step.
Is Nw Roadmap Review Checks safe to install?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean skill reads B.
What can Nw Roadmap Review Checks access on my machine?
The audit observed no filesystem, network or shell use at all in its source.
How current is this page?
The grade is for one exact copy of the source (1a864ec81cde), read on 2026-10-14. The repository is watched, and a new audit runs when it changes — this is the first audit.