feat: AGT-inspired governance improvements — severity registry, trust tiers, limitations doc, conformance manifest - #48
Merged
Conversation
Adds three new governance companion documents and updates the constitution:
- severity-registry.md: maps §I–§VIII to §{section}.{LEVEL} finding codes
- known-limitations.md: six honest governance scope boundaries (L-001–L-006)
- prompt-conformance-manifest.md: deterministic conformance rules for command templates
- constitution.md: Companion Documents section linking all three artifacts
Part of spec 001-agt-governance-improvements (FR-001, FR-005, FR-007).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Updates to pr-review.md:
- Step 1b: trust-tier classification (full/partial/no-compliance detection)
- Step 2: load severity-registry.md for finding code validation
- Guidelines: severity code format §{section}.{LEVEL} mandate
Updates to evolve-constitution.md:
- Review Checklist: severity-registry co-update item (FR-009)
- Review Checklist: known-limitations check item (FR-006)
CLAUDE.md: fix pre-existing markdownlint errors from agent context update.
Part of spec 001-agt-governance-improvements (FR-002, FR-003, FR-004, FR-006, FR-009).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Full spec-driven development artifacts for AGT-inspired governance improvements: - spec.md (Complete), plan.md, tasks.md (16/16 done), research.md - checklists/requirements.md - gates/analyze.md (pass), gates/critic.md (pass) All 16 tasks complete, both pre-implement gates resolved, markdownlint 0 errors across 381 files. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
markhazleton
force-pushed
the
001-agt-governance-improvements
branch
from
June 3, 2026 23:44
9bdf318 to
1cfd3c0
Compare
H-01: Move trust-tier⚠️ alert blockquote inside the no-compliance conditional (step 5) and add explicit partial-compliance inline note. Fixes §II Explicit Over Implied violation — alert no longer fires unconditionally for partial-compliance branches. M-01: Remove blank lines inside HTML comment block in prompt-conformance-manifest.md baseline section. L-01: Use relative path 'constitution.md' (not '.documentation/memory/constitution.md') in severity-registry.md Companion Documents section. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
H-01 resolved: trust-tier alert now conditional on no-compliance only. M-01 resolved: HTML comment blank lines removed. L-01 resolved: relative path in severity-registry companion links. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Adds four governance improvements to DevSpark inspired by Microsoft's Agent Governance Toolkit philosophy — structured severity codes, trust-tiered review depth, explicit limitations documentation, and prompt conformance checking — all as additive Markdown artifacts with no new tool dependencies.
severity-registry.md): maps §I–§VIII constitution sections to§{section}.{LEVEL}finding codes;pr-reviewnow loads and emits these codespr-reviewclassifies branches as full/partial/no-compliance and emits a MEDIUM finding + reviewer alert for no-spec branchesknown-limitations.md): six honest scope boundaries (L-001–L-006) alongside the constitutionprompt-conformance-manifest.md): deterministic string-match rules for validating command templates via/devspark.checklistevolve-constitution.mdupdated: two new Review Checklist items enforce registry and limitations sync on every amendment (FR-006, FR-009)Changes
13 files changed, 1644 insertions(+)
.documentation/memory/severity-registry.md.documentation/memory/known-limitations.md.documentation/memory/prompt-conformance-manifest.md.documentation/memory/constitution.mdtemplates/commands/pr-review.mdtemplates/commands/evolve-constitution.mdCLAUDE.md(pre-existing lint fix).documentation/specs/001-agt-governance-improvements/spec.md.documentation/specs/001-agt-governance-improvements/plan.md.documentation/specs/001-agt-governance-improvements/tasks.md.documentation/specs/001-agt-governance-improvements/research.md.documentation/specs/001-agt-governance-improvements/gates/analyze.md.documentation/specs/001-agt-governance-improvements/gates/critic.mdTask Completion
16/16 tasks complete | Checklist: 25/25
Quality Gates
Spec Reference
.documentation/specs/001-agt-governance-improvements/spec.md—full-spec|mediumrisk | Status: CompleteNotes
pr-review.mdis additive only (no content removed)CLAUDE.mdfix: corrected pre-existing H1 heading and trailing blank line introduced by agent context update script🤖 Generated with Claude Code