CtrlK
BlogDocsLog inGet started
Tessl Logo

sporkwace/guv

enforces engineering-governance checks before code changes that may be unnecessary, risky, architectural, or scope-widening. use when the user asks whether to refactor, clean up, redesign, choose a next development step, review a proposed implementation, evaluate architectural consistency, review a pull request, or prevent development drift. do not use as the primary implementation skill for routine debugging, bug fixes, feature coding, or language-specific coding unless a no-op, minimal-diff, or architecture-conflict judgment is needed.

98

1.33x
Quality

100%

Does it follow best practices?

Impact

96%

1.33x

Average score across 5 eval scenarios

SecuritybySnyk

Passed

No known issues

Overview
Quality
Evals
Security
Files

Evaluation results

100%

31%

Should we touch the formatter?

Criteria
Without context
With context

Purpose

100%

100%

No benefit

60%

100%

No-op check

40%

100%

Architecture impact

20%

100%

Risks

100%

100%

Smallest change

50%

100%

Acceptance test

70%

100%

Recommendation

80%

100%

No-op value

70%

100%

Style pushback

100%

100%

98%

6%

Receipt Formatter Change

Criteria
Without context
With context

Concrete reason

90%

100%

Smallest diff

100%

100%

Diff only if useful

90%

100%

Relevant files

90%

100%

Behavior preserved

100%

100%

Focused tests

90%

100%

Existing patterns

70%

80%

Explain necessity

90%

100%

No aesthetics

100%

100%

Scope discipline

100%

100%

100%

30%

Direct Analytics Write Proposal

Criteria
Without context
With context

Context

100%

100%

Decision

100%

100%

Alternatives

100%

100%

Constraints

10%

100%

Consequences

20%

100%

Acceptance test

0%

100%

Source of truth

100%

100%

Write boundary

100%

100%

Duplication risk

100%

100%

Traceability

70%

100%

88%

19%

Checkout Next Step

Criteria
Without context
With context

Option comparison

50%

70%

Why now

70%

70%

Conflicts

50%

90%

Deferrals

80%

90%

Scope impact

40%

90%

Acceptance test

80%

90%

Recommendation

100%

100%

Stabilization bias

90%

100%

Project goal

70%

80%

Governance frame

60%

100%

94%

34%

Modernization PR Review

Criteria
Without context
With context

Churn

90%

100%

Duplication

20%

100%

Ownership

20%

100%

Abstraction

60%

100%

Drift

20%

80%

Untested

30%

60%

Scope

80%

100%

Pushback

90%

100%

Objection

90%

100%

Safer alternative

100%

100%

Evaluated
Agent
Claude
Model
Claude Sonnet 4.6

Table of Contents