LLM-Coding · rdmueller · Mar 14, 2026 · Mar 14, 2026
diff --git a/.github/ISSUE_TEMPLATE/propose-anchor.yml b/.github/ISSUE_TEMPLATE/propose-anchor.yml
@@ -11,12 +11,33 @@ body:
         Thank you for proposing a new semantic anchor! Here's the workflow:
 
         1. **You provide a term or concept** that you believe would make a good semantic anchor
-        2. **GitHub Copilot validates** whether it meets the quality criteria (precise, rich, consistent, attributable)
-        3. **If accepted**: Copilot enriches this issue with detailed information (proponents, categories, related concepts)
-        4. **If rejected**: Copilot explains why it doesn't meet the criteria
-        5. **After enrichment**: Copilot is assigned to create the anchor file and submit a PR
+        2. **The review agent validates** whether it meets the quality criteria (see below)
+        3. **If accepted**: The agent enriches this issue with detailed information (proponents, categories, related concepts) and assigns a quality tier
+        4. **If rejected**: The agent explains why it doesn't meet the criteria
+        5. **After enrichment**: The agent creates the anchor file and submits a PR
 
-        **Just provide the term below - Copilot handles the rest!**
+        ### Quality Criteria
+
+        A semantic anchor must be **precise**, **rich**, **consistent**, and **attributable**.
+
+        ### Tier Rating (★)
+
+        The review agent assigns one of three tiers based on how reliably the anchor improves LLM communication:
+
+        - **★★★ Self-standing**: Saying the anchor name alone reliably activates the correct behavior in the LLM. Clear, actionable, deterministic output. Examples: arc42, SOLID, Gherkin, MECE, Chain of Thought.
+        - **★★☆ Needs qualification**: The concept is known but requires a qualifier for reliable results. The anchor documentation must specify the recommended qualified form. Examples: "ADR according to Nygard", "3-point Pugh Matrix".
+        - **★☆☆ Descriptive only**: The term describes a concept but does not give the LLM clear instructions. These are **rejected** unless the proposer can demonstrate a concrete, actionable prompt pattern.
+
+        ### Rejection Criteria
+
+        The review agent **rejects** proposals that are:
+        - Pure philosophy without actionable instructions (e.g., "TIMTOWTDI")
+        - Personality/assessment frameworks without LLM utility (e.g., "MBTI")
+        - Fully covered by an existing anchor (e.g., "Rubber Duck Debugging" ≈ Socratic Method)
+        - Too vague to produce consistent LLM output (e.g., "Keep it simple")
+        - Not an established concept (invented by the proposer)
+
+        **Just provide the term below - the review agent handles the rest!**
 
   - type: input
     id: term

diff --git a/docs/anchors/_template.adoc b/docs/anchors/_template.adoc
@@ -4,6 +4,7 @@
 :related: related-anchor-1, related-anchor-2
 :proponents: Author Name, Another Author
 :tags: tag1, tag2, tag3
+:tier: 3
 
 [%collapsible]
 ====

diff --git a/docs/anchors/adr-according-to-nygard.adoc b/docs/anchors/adr-according-to-nygard.adoc
@@ -1,6 +1,7 @@
 = ADR according to Nygard
 :categories: software-architecture
 :roles: software-architect, software-developer, team-lead
+:tier: 3
 
 [%collapsible]
 ====

diff --git a/docs/anchors/arc42.adoc b/docs/anchors/arc42.adoc
@@ -2,6 +2,7 @@
 :categories: software-architecture
 :roles: software-architect, technical-writer, team-lead
 :proponents: Gernot Starke, Peter Hruschka
+:tier: 3
 
 [%collapsible]
 ====

diff --git a/docs/anchors/atam.adoc b/docs/anchors/atam.adoc
@@ -4,6 +4,7 @@
 :related: arc42, adr-according-to-nygard
 :proponents: Rick Kazman, Mark Klein, Paul Clements
 :tags: architecture evaluation, quality attributes, tradeoffs, utility tree, SEI, risk analysis
+:tier: 3
 
 [%collapsible]
 ====

diff --git a/docs/anchors/bdd-given-when-then.adoc b/docs/anchors/bdd-given-when-then.adoc
@@ -4,6 +4,7 @@
 :proponents: Dan North
 :related: tdd-london-school, tdd-chicago-school, user-story-mapping
 :tags: bdd, gherkin, cucumber, specification-by-example, three-amigos, given-when-then, living-documentation
+:tier: 3
 
 [%collapsible]
 ====

diff --git a/docs/anchors/bem-methodology.adoc b/docs/anchors/bem-methodology.adoc
@@ -2,6 +2,7 @@
 :categories: development-workflow
 :roles: software-developer, ux-designer
 :proponents: Yandex development team
+:tier: 3
 
 [%collapsible]
 ====

diff --git a/docs/anchors/bluf.adoc b/docs/anchors/bluf.adoc
@@ -2,6 +2,7 @@
 :categories: communication-presentation
 :roles: business-analyst, team-lead, technical-writer, consultant
 :proponents: US Military (Army writing standards), adopted broadly in business and government
+:tier: 3
 
 [%collapsible]
 ====

diff --git a/docs/anchors/c4-diagrams.adoc b/docs/anchors/c4-diagrams.adoc
@@ -1,6 +1,7 @@
 = C4-Diagrams
 :categories: software-architecture
 :roles: software-architect, technical-writer, team-lead, consultant
+:tier: 3
 
 [%collapsible]
 ====

diff --git a/docs/anchors/chain-of-thought.adoc b/docs/anchors/chain-of-thought.adoc
@@ -2,6 +2,7 @@
 :categories: problem-solving
 :roles: software-developer, data-scientist, business-analyst
 :proponents: Wei et al. (Google Research, 2022), "Chain-of-Thought Prompting Elicits Reasoning in Large Language Models"
+:tier: 3
 
 [%collapsible]
 ====

diff --git a/docs/anchors/chatham-house-rule.adoc b/docs/anchors/chatham-house-rule.adoc
diff --git a/docs/anchors/chatham-house-rule.de.adoc b/docs/anchors/chatham-house-rule.de.adoc
diff --git a/docs/anchors/clean-architecture.adoc b/docs/anchors/clean-architecture.adoc
@@ -1,6 +1,7 @@
 = Clean Architecture
 :categories: software-architecture
 :roles: software-architect, software-developer
+:tier: 3
 
 [%collapsible]
 ====

diff --git a/docs/anchors/control-chart-shewhart.adoc b/docs/anchors/control-chart-shewhart.adoc
@@ -1,6 +1,8 @@
 = Control Chart (Shewhart)
 :categories: statistical-methods
 :roles: data-scientist, devops-engineer, team-lead
+:tier: 3
+:parent-anchor: spc
 
 [%collapsible]
 ====

diff --git a/docs/anchors/conventional-commits.adoc b/docs/anchors/conventional-commits.adoc
@@ -2,6 +2,7 @@
 :categories: development-workflow
 :roles: software-developer, devops-engineer
 :proponents: Benjamin E. Coe, James J. Womack, Steve Mao
+:tier: 3
 
 [%collapsible]
 ====

diff --git a/docs/anchors/cqrs.adoc b/docs/anchors/cqrs.adoc
@@ -4,6 +4,7 @@
 :proponents: Greg Young, Bertrand Meyer, Udi Dahan
 :related: domain-driven-design, hexagonal-architecture, fowler-patterns
 :tags: cqrs, cqs, architecture, commands, queries, read-write-separation
+:tier: 3
 
 [%collapsible]
 ====

diff --git a/docs/anchors/cynefin-framework.adoc b/docs/anchors/cynefin-framework.adoc
@@ -4,6 +4,7 @@
 :proponents: Dave Snowden
 :related: wardley-mapping
 :tags: complexity, decision-making, sensemaking, strategy, domains
+:tier: 3
 
 [%collapsible]
 ====

diff --git a/docs/anchors/devils-advocate.adoc b/docs/anchors/devils-advocate.adoc
@@ -1,6 +1,7 @@
 = Devil's Advocate
 :categories: problem-solving
 :roles: software-architect, team-lead, consultant, qa-engineer
+:tier: 3
 
 [%collapsible]
 ====

diff --git a/docs/anchors/diataxis-framework.adoc b/docs/anchors/diataxis-framework.adoc
@@ -1,6 +1,7 @@
 = Diátaxis Framework
 :categories: documentation
 :roles: technical-writer, educator
+:tier: 3
 
 [%collapsible]
 ====

diff --git a/docs/anchors/docs-as-code.adoc b/docs/anchors/docs-as-code.adoc
@@ -1,6 +1,7 @@
 = Docs-as-Code according to Ralf D. Müller
 :categories: documentation
 :roles: technical-writer, software-developer, devops-engineer
+:tier: 3
 
 [%collapsible]
 ====

diff --git a/docs/anchors/domain-driven-design.adoc b/docs/anchors/domain-driven-design.adoc
@@ -1,6 +1,7 @@
 = Domain-Driven Design according to Evans
 :categories: software-architecture
 :roles: software-architect, software-developer, business-analyst, team-lead
+:tier: 3
 
 [%collapsible]
 ====

diff --git a/docs/anchors/dry-principle.adoc b/docs/anchors/dry-principle.adoc