ai-factory 2.2.2 → 2.4.0

package/skills/aif-explore/SKILL.md

@@ -0,0 +1,384 @@
+---
+name: aif-explore
+description: Enter explore mode - a thinking partner for exploring ideas, investigating problems, and clarifying requirements. Use when the user wants to think through something before or during a change.
+argument-hint: "[topic or plan name]"
+allowed-tools: Read Glob Grep Write Edit Bash AskUserQuestion Questions
+disable-model-invocation: true
+---
+
+Enter explore mode. Think deeply. Visualize freely. Follow the conversation wherever it goes.
+
+**IMPORTANT: Explore mode is for thinking, not implementing.** You may read files, search code, and investigate the codebase, but you must NEVER implement features or modify project code. If the user asks to implement something, remind them to exit explore mode first (e.g., start with `/aif-plan`). If the user asks to persist exploration context, write/edit **only** `.ai-factory/RESEARCH.md` (this is capturing thinking, not implementing).
+
+**This is a stance, not a workflow.** There are no fixed steps, no required sequence, no mandatory outputs. You're a thinking partner helping the user explore.
+
+---
+
+## Artifact Ownership
+
+- Primary ownership in explore mode: `.ai-factory/RESEARCH.md` only.
+- All other context artifacts (`DESCRIPTION.md`, `ARCHITECTURE.md`, `ROADMAP.md`, `RULES.md`, plan files) are read-only in this mode.
+- If a discovery should affect another artifact, capture it in RESEARCH now and route follow-up to the owner command later.
+
+---
+
+## The Stance
+
+- **Curious, not prescriptive** - Ask questions that emerge naturally, don't follow a script
+- **Open threads, not interrogations** - Surface multiple interesting directions and let the user follow what resonates. Don't funnel them through a single path of questions.
+- **Visual** - Use ASCII diagrams liberally when they'd help clarify thinking
+- **Adaptive** - Follow interesting threads, pivot when new information emerges
+- **Patient** - Don't rush to conclusions, let the shape of the problem emerge
+- **Grounded** - Explore the actual codebase when relevant, don't just theorize
+
+---
+
+## What You Might Do
+
+Depending on what the user brings, you might:
+
+**Explore the problem space**
+- Ask clarifying questions that emerge from what they said
+- Challenge assumptions
+- Reframe the problem
+- Find analogies
+
+**Investigate the codebase**
+- Map existing architecture relevant to the discussion
+- Find integration points
+- Identify patterns already in use
+- Surface hidden complexity
+
+**Compare options**
+- Brainstorm multiple approaches
+- Build comparison tables
+- Sketch tradeoffs
+- Recommend a path (if asked)
+
+**Visualize**
+```
++-----------------------------------------+
+|     Use ASCII diagrams liberally        |
++-----------------------------------------+
+|                                         |
+|   +--------+         +--------+        |
+|   | State  |-------->| State  |        |
+|   |   A    |         |   B    |        |
+|   +--------+         +--------+        |
+|                                         |
+|   System diagrams, state machines,      |
+|   data flows, architecture sketches,    |
+|   dependency graphs, comparison tables  |
+|                                         |
++-----------------------------------------+
+```
+
+**Surface risks and unknowns**
+- Identify what could go wrong
+- Find gaps in understanding
+- Suggest spikes or investigations
+
+---
+
+## AI Factory Context
+
+You have access to AI Factory's project context. Use it naturally, don't force it.
+
+**Read `.ai-factory/skill-context/aif-explore/SKILL.md`** — MANDATORY if the file exists.
+
+This file contains project-specific rules accumulated by `/aif-evolve` from patches,
+codebase conventions, and tech-stack analysis. These rules are tailored to the current project.
+
+**How to apply skill-context rules:**
+- Treat them as **project-level overrides** for this skill's general instructions
+- When a skill-context rule conflicts with a general rule written in this SKILL.md,
+  **the skill-context rule wins** (more specific context takes priority — same principle as nested CLAUDE.md files)
+- When there is no conflict, apply both: general rules from SKILL.md + project rules from skill-context
+- Do NOT ignore skill-context rules even if they seem to contradict this skill's defaults —
+  they exist because the project's experience proved the default insufficient
+- **CRITICAL:** skill-context rules apply to ALL outputs of this skill — including exploration
+  summaries, diagrams, and any file updates (DESCRIPTION.md, ARCHITECTURE.md). If a skill-context
+  rule says "exploration MUST cover X" or "summary MUST include Y" — you MUST comply. Producing
+  output that ignores skill-context rules is a bug.
+
+**Enforcement:** After generating any output artifact, verify it against all skill-context rules.
+If any rule is violated — fix the output before presenting it to the user.
+
+### Check for context
+
+At the start, read these files if present:
+
+- `.ai-factory/DESCRIPTION.md` — project description, tech stack, constraints
+- `.ai-factory/ARCHITECTURE.md` — architecture decisions, folder structure
+- `.ai-factory/RULES.md` — project conventions and rules
+- `.ai-factory/RESEARCH.md` — persisted exploration notes (so you can `/clear` and still keep context)
+- `.ai-factory/PLAN.md` — active fast plan (if any)
+- `.ai-factory/plans/<branch>.md` — active full plans (if any)
+- `.ai-factory/ROADMAP.md` — strategic milestones (if any)
+
+This tells you:
+- What the project is about
+- What conventions to follow
+- If there's active work in progress
+- Any prior exploration context worth carrying into planning
+
+### Input handling
+
+The argument after `/aif-explore` can be:
+- A vague idea: "real-time collaboration"
+- A specific problem: "the auth system is getting unwieldy"
+- A plan name: to explore in context of `.ai-factory/plans/<name>.md`
+- A comparison: "postgres vs sqlite for this"
+- Nothing: just enter explore mode
+
+### When no plan exists
+
+Think freely. When insights crystallize, you might offer:
+
+- "This feels solid enough to plan. Want me to start `/aif-plan`?"
+- Or keep exploring - no pressure to formalize
+
+### When a plan exists
+
+If the user mentions a plan or you detect one is relevant:
+
+1. **Read existing plan for context**
+   - `.ai-factory/PLAN.md` (fast mode)
+   - `.ai-factory/plans/<branch>.md` (full mode)
+
+2. **Reference it naturally in conversation**
+   - "Your plan mentions adding Redis, but we just realized SQLite fits better..."
+   - "Task 3 scopes this to premium users, but we're now thinking everyone..."
+
+3. **Offer to capture when decisions are made**
+
+   Default in explore mode: capture everything in `.ai-factory/RESEARCH.md` so it survives `/clear`.
+   Later (during planning), you can migrate stabilized decisions into the appropriate context file.
+
+   | Insight Type | Capture Now (Explore) | Later (Optional) |
+   |--------------|------------------------|------------------|
+   | New requirement | `.ai-factory/RESEARCH.md` | `.ai-factory/DESCRIPTION.md` |
+   | Architecture decision | `.ai-factory/RESEARCH.md` | `.ai-factory/ARCHITECTURE.md` |
+   | Project convention | `.ai-factory/RESEARCH.md` | `.ai-factory/RULES.md` |
+   | Strategic direction | `.ai-factory/RESEARCH.md` | `.ai-factory/ROADMAP.md` |
+   | Assumption invalidated | `.ai-factory/RESEARCH.md` | Relevant file |
+   | Exploration context (persisted) | `.ai-factory/RESEARCH.md` | (keep in RESEARCH) |
+   | New task/feature | Run `/aif-plan` | `.ai-factory/PLAN.md` or `.ai-factory/plans/<branch>.md` |
+
+   Example offers:
+   - "Want me to save this to `.ai-factory/RESEARCH.md` so you can `/clear` and come back later?"
+   - "That's an architecture decision — save it to RESEARCH now and we can migrate it to ARCHITECTURE during planning."
+
+4. **The user decides** - Offer and move on. Don't pressure. Don't auto-capture.
+
+### Optional: Persist exploration context (`.ai-factory/RESEARCH.md`)
+
+If the conversation is crystallizing (you're about to plan, you want to `/clear`, or you want to continue later), offer to save a compact, durable research snapshot.
+
+**Hard rule in explore mode:** If the user chooses to save, you may write/edit **only** `.ai-factory/RESEARCH.md` (and create the `.ai-factory/` directory if missing). Do not write or modify any other project files.
+
+Ask:
+
+```
+Save these exploration results to .ai-factory/RESEARCH.md so we can /clear and /aif-plan can reuse them?
+
+Options:
+1. Yes — update Active Summary + append a new Session (recommended)
+2. Yes — update Active Summary only
+3. No
+```
+
+If user selects (1) or (2):
+- Ensure `.ai-factory/` exists (`mkdir -p .ai-factory`)
+- If `.ai-factory/RESEARCH.md` does not exist, create it with this skeleton:
+
+```markdown
+# Research
+
+Updated: YYYY-MM-DD HH:MM
+Status: active
+
+## Active Summary (input for /aif-plan)
+<!-- aif:active-summary:start -->
+Topic:
+Goal:
+Constraints:
+Decisions:
+Open questions:
+Success signals:
+Next step:
+<!-- aif:active-summary:end -->
+
+## Sessions
+<!-- aif:sessions:start -->
+<!-- aif:sessions:end -->
+```
+
+- Update the `Updated:` timestamp
+- Replace only the content inside `aif:active-summary:start/end`
+- If user selected option (1), append a new session entry just before `<!-- aif:sessions:end -->`:
+
+```markdown
+### YYYY-MM-DD HH:MM — <short title>
+What changed:
+Key notes:
+Links (paths):
+```
+
+Keep prior sessions verbatim (do not rewrite history).
+
+---
+
+## What You Don't Have To Do
+
+- Follow a script
+- Ask the same questions every time
+- Produce a specific artifact
+- Reach a conclusion
+- Stay on topic if a tangent is valuable
+- Be brief (this is thinking time)
+
+---
+
+## Handling Different Entry Points
+
+**User brings a vague idea:**
+```
+User: I'm thinking about adding real-time collaboration
+
+You: Real-time collab is a big space. Let me think about this...
+
+      COLLABORATION SPECTRUM
+
+      Awareness          Coordination         Sync
+          |                   |                 |
+          v                   v                 v
+      +--------+         +--------+        +--------+
+      |Presence|         |Cursors |        |  CRDT  |
+      |  "3    |         | Multi  |        |Conflict|
+      |online" |         | select |        |  free  |
+      +--------+         +--------+        +--------+
+          |                   |                 |
+       trivial            moderate           complex
+
+      Where's your head at?
+```
+
+**User brings a specific problem:**
+```
+User: The auth system is a mess
+
+You: [reads codebase]
+
+     +---------------------------------------------+
+     |              CURRENT AUTH FLOW              |
+     +---------------------------------------------+
+                        |
+          +-------------+-------------+
+          v             v             v
+     +---------+  +---------+  +---------+
+     | Google  |  |  GitHub |  |  Email  |
+     |  OAuth  |  |  OAuth  |  |  Magic  |
+     +----+----+  +----+----+  +----+----+
+          |             |             |
+          +-------------+-------------+
+                        v
+                  +-----------+
+                  |  Session  |
+                  +-----+-----+
+                        |
+                        v
+                  +-----------+
+                  |   Perms   |
+                  +-----------+
+
+     I see three tangles. Which one's burning?
+```
+
+**User is stuck mid-implementation:**
+```
+User: /aif-explore add-auth-system
+      The OAuth integration is more complex than expected
+
+You: [reads plan from .ai-factory/plans/add-auth-system.md]
+
+     You're on task 4: "Implement OAuth flow"
+
+     Let me trace what's involved...
+
+     [draws diagram, explores options, suggests paths]
+
+     Want to update the plan to reflect this?
+     Or add a spike task to investigate?
+```
+
+**User wants to compare options:**
+```
+User: Should we use Postgres or SQLite?
+
+You: Generic answer is boring. What's the context?
+
+User: A CLI tool that tracks local dev environments
+
+You: That changes everything.
+
+     +---------------------------------------------+
+     |         CLI TOOL DATA STORAGE               |
+     +---------------------------------------------+
+
+     Key constraints:
+     * No daemon running
+     * Must work offline
+     * Single user
+
+                  SQLite          Postgres
+     Deployment   embedded ok     needs server no
+     Offline      yes ok          no no
+     Single file  yes ok          no no
+
+     SQLite. Not even close.
+
+     Unless... is there a sync component?
+```
+
+---
+
+## Ending Discovery
+
+There's no required ending. Discovery might:
+
+- **Flow into action**: "Ready to plan? Run `/aif-plan`"
+- **Result in context updates**: "Updated ARCHITECTURE.md with these decisions"
+- **Just provide clarity**: User has what they need, moves on
+- **Continue later**: "We can pick this up anytime"
+
+When it feels like things are crystallizing, you might summarize:
+
+```
+## What We Figured Out
+
+**The problem**: [crystallized understanding]
+
+**The approach**: [if one emerged]
+
+**Open questions**: [if any remain]
+
+**Next steps** (if ready):
+- Create a plan: /aif-plan [fast|full] <description>
+- Keep exploring: just keep talking
+```
+
+But this summary is optional. Sometimes the thinking IS the value.
+
+---
+
+## Guardrails
+
+- **Don't implement** - Never write code or implement features. Updating AI Factory context files is fine, writing application code is not.
+- **Don't fake understanding** - If something is unclear, dig deeper
+- **Don't rush** - Discovery is thinking time, not task time
+- **Don't force structure** - Let patterns emerge naturally
+- **Don't auto-capture** - Offer to save insights, don't just do it
+- **Do visualize** - A good diagram is worth many paragraphs
+- **Do explore the codebase** - Ground discussions in reality
+- **Do question assumptions** - Including the user's and your own

package/skills/aif-fix/SKILL.md

@@ -48,6 +48,26 @@ Fix a specific bug or problem in the codebase. Supports two modes: immediate fix
 - If the current problem resembles a past patch — apply the same approach or avoid the same mistakes
 - This is your accumulated experience. Use it.
 
+**Read `.ai-factory/skill-context/aif-fix/SKILL.md`** — MANDATORY if the file exists.
+
+This file contains project-specific rules accumulated by `/aif-evolve` from patches,
+codebase conventions, and tech-stack analysis. These rules are tailored to the current project.
+
+**How to apply skill-context rules:**
+- Treat them as **project-level overrides** for this skill's general instructions
+- When a skill-context rule conflicts with a general rule written in this SKILL.md,
+  **the skill-context rule wins** (more specific context takes priority — same principle as nested CLAUDE.md files)
+- When there is no conflict, apply both: general rules from SKILL.md + project rules from skill-context
+- Do NOT ignore skill-context rules even if they seem to contradict this skill's defaults —
+  they exist because the project's experience proved the default insufficient
+- **CRITICAL:** skill-context rules apply to ALL outputs of this skill — including the FIX_PLAN.md
+  template and patch files. The FIX_PLAN.md template in Step 1.1 is a **base structure**. If a
+  skill-context rule says "steps MUST include X" or "plan MUST have section Y" — you MUST augment
+  the template accordingly. Generating a FIX_PLAN.md or patch that violates skill-context rules is a bug.
+
+**Enforcement:** After generating any output artifact, verify it against all skill-context rules.
+If any rule is violated — fix the output before presenting it to the user.
+
 ### Step 1: Understand the Problem & Choose Mode
 
 From `$ARGUMENTS`, identify:
@@ -324,6 +344,8 @@ function fixedFunction(input) {
 8. **Minimal changes** - Don't refactor unrelated code
 9. **One fix at a time** - Don't scope creep
 10. **Clean up** - Delete FIX_PLAN.md after successful fix execution
+11. **Ownership boundary** - `/aif-fix` owns `.ai-factory/FIX_PLAN.md` and `.ai-factory/patches/*.md`; treat `.ai-factory/DESCRIPTION.md`, roadmap/rules/architecture context artifacts as read-only unless the user explicitly requests otherwise
+12. **Logging scope** - Keep `[FIX]` logging requirements for fixes; context-gate outputs in this command should use `WARN`/`ERROR` and must not change global logging policy in other skills
 
 ## After Fixing
 

package/skills/aif-grounded/SKILL.md

@@ -24,6 +24,28 @@ Use when:
 
 ## Workflow
 
+### Step 0: Load Skill Context
+
+**Read `.ai-factory/skill-context/aif-grounded/SKILL.md`** — MANDATORY if the file exists.
+
+This file contains project-specific rules accumulated by `/aif-evolve` from patches,
+codebase conventions, and tech-stack analysis. These rules are tailored to the current project.
+
+**How to apply skill-context rules:**
+- Treat them as **project-level overrides** for this skill's general instructions
+- When a skill-context rule conflicts with a general rule written in this SKILL.md,
+  **the skill-context rule wins** (more specific context takes priority — same principle as nested CLAUDE.md files)
+- When there is no conflict, apply both: general rules from SKILL.md + project rules from skill-context
+- Do NOT ignore skill-context rules even if they seem to contradict this skill's defaults —
+  they exist because the project's experience proved the default insufficient
+- **CRITICAL:** skill-context rules apply to ALL outputs of this skill — including the response
+  format, evidence requirements, and confidence assessment. If a skill-context rule says "analysis
+  MUST include X" or "confidence MUST account for Y" — you MUST comply. Producing an analysis
+  that ignores skill-context rules is a bug.
+
+**Enforcement:** After generating any output artifact, verify it against all skill-context rules.
+If any rule is violated — fix the output before presenting it to the user.
+
 ### Step 1: Classify the request
 
 Classify into one of:

package/skills/aif-implement/SKILL.md

@@ -3,7 +3,7 @@ name: aif-implement
 description: Execute implementation tasks from the current plan. Works through tasks sequentially, marks completion, and preserves progress for continuation across sessions. Use when user says "implement", "start coding", "execute plan", or "continue implementation".
 argument-hint: '[task-id or "status"]'
 allowed-tools: Read Write Edit Glob Grep Bash TaskList TaskGet TaskUpdate AskUserQuestion Questions
-disable-model-invocation: true
+disable-model-invocation: false
 ---
 
 # Implement - Execute Task Plan
@@ -115,6 +115,26 @@ Based on choice:
 - Apply proper error handling and logging as specified
 - **Avoid pitfalls documented in patches** — don't repeat past mistakes
 
+**Read `.ai-factory/skill-context/aif-implement/SKILL.md`** — MANDATORY if the file exists.
+
+This file contains project-specific rules accumulated by `/aif-evolve` from patches,
+codebase conventions, and tech-stack analysis. These rules are tailored to the current project.
+
+**How to apply skill-context rules:**
+- Treat them as **project-level overrides** for this skill's general instructions
+- When a skill-context rule conflicts with a general rule written in this SKILL.md,
+  **the skill-context rule wins** (more specific context takes priority — same principle as nested CLAUDE.md files)
+- When there is no conflict, apply both: general rules from SKILL.md + project rules from skill-context
+- Do NOT ignore skill-context rules even if they seem to contradict this skill's defaults —
+  they exist because the project's experience proved the default insufficient
+- **CRITICAL:** skill-context rules apply to ALL outputs of this skill — including the code
+  you write and how you update plan checkboxes. If a skill-context rule says "code MUST follow X"
+  or "implementation MUST include Y" — you MUST comply. Writing code that violates skill-context
+  rules is a bug.
+
+**Enforcement:** After generating any output artifact, verify it against all skill-context rules.
+If any rule is violated — fix the output before presenting it to the user.
+
 ### Step 0.1: Find Plan File
 
 **Check for plan files in this order:**
@@ -302,10 +322,44 @@ What's next?
 
 If `.ai-factory/ROADMAP.md` exists:
 1. Read it
+1.1. If the plan file includes `## Roadmap Linkage` with a non-`none` milestone, prefer that milestone for completion marking
 2. Check if the completed work corresponds to any unchecked milestone
 3. If yes — mark it `[x]` and add entry to the Completed table with today's date
 4. Tell the user which milestone was marked done
 
+### Context Maintenance (Artifacts)
+
+Only do this step when there is something concrete to capture.
+
+**DESCRIPTION.md (allowed in this command):**
+- If this plan introduced new dependencies/integrations or changed the stack, update `.ai-factory/DESCRIPTION.md` with factual deltas only.
+- Do not rewrite unrelated sections.
+
+**ARCHITECTURE.md + AGENTS.md (allowed in this command):**
+- If new modules/layers/folders were added (or dependency rules changed), update `.ai-factory/ARCHITECTURE.md` to reflect the new structure and constraints.
+- If you maintain `AGENTS.md` structure maps or entry points, refresh them only when they are now incorrect.
+
+**ROADMAP.md (allowed, limited):**
+- This command may mark milestone completion when evidence is clear.
+- If milestone mapping is ambiguous, emit `WARN [roadmap] ...` and suggest the owner command:
+  - `/aif-roadmap check`
+  - or `/aif-roadmap <short update request>`
+
+**RULES.md (NOT allowed in this command):**
+- Never edit `.ai-factory/RULES.md` from `/aif-implement`.
+- If you discovered repeatable conventions/pitfalls during implementation, propose up to 3 candidate rules and ask the user to add them via `/aif-rules`.
+- Do not invoke `/aif-rules` automatically (it is user-invoked).
+
+If candidate rules exist:
+
+```
+AskUserQuestion: Capture new project rules in `.ai-factory/RULES.md`?
+
+Options:
+1. Yes — output `/aif-rules ...` commands (recommended)
+2. No — skip
+```
+
 ### Context Cleanup
 
 Context is heavy after implementation. All code changes are saved — suggest freeing space:
@@ -477,6 +531,13 @@ Shows progress without executing.
 - ❌ Mark incomplete tasks as done
 - ❌ Violate `.ai-factory/ARCHITECTURE.md` conventions for file placement and module boundaries
 
+## Artifact Ownership Boundaries
+
+- Primary ownership in this command: task execution state and plan progress checkboxes.
+- Allowed context artifact updates: `.ai-factory/DESCRIPTION.md`, `.ai-factory/ARCHITECTURE.md`, and roadmap milestone completion in `.ai-factory/ROADMAP.md` when implementation evidence justifies it.
+- Read-only context in this command by default: `.ai-factory/RULES.md`, `.ai-factory/RESEARCH.md`.
+- Context-gate findings should be communicated as `WARN`/`ERROR` outputs only; this does not replace the required verbose implementation logging rules below.
+
 For progress display format, blocker handling, session continuity examples, and full flow examples → see `references/IMPLEMENTATION-GUIDE.md`
 
 ## Critical Rules

package/skills/aif-improve/SKILL.md

@@ -79,6 +79,26 @@ If patches exist, read them to understand:
 - What patterns to avoid
 - What the plan should account for
 
+**Read `.ai-factory/skill-context/aif-improve/SKILL.md`** — MANDATORY if the file exists.
+
+This file contains project-specific rules accumulated by `/aif-evolve` from patches,
+codebase conventions, and tech-stack analysis. These rules are tailored to the current project.
+
+**How to apply skill-context rules:**
+- Treat them as **project-level overrides** for this skill's general instructions
+- When a skill-context rule conflicts with a general rule written in this SKILL.md,
+  **the skill-context rule wins** (more specific context takes priority — same principle as nested CLAUDE.md files)
+- When there is no conflict, apply both: general rules from SKILL.md + project rules from skill-context
+- Do NOT ignore skill-context rules even if they seem to contradict this skill's defaults —
+  they exist because the project's experience proved the default insufficient
+- **CRITICAL:** skill-context rules apply to ALL outputs of this skill — including the Plan
+  Refinement Report and any plan modifications. If a skill-context rule says "tasks MUST include X"
+  or "plan structure MUST have Y" — you MUST apply these when refining. Generating a refinement
+  report that ignores skill-context rules is a bug.
+
+**Enforcement:** After generating any output artifact, verify it against all skill-context rules.
+If any rule is violated — fix the output before presenting it to the user.
+
 **1.4: Load current task list**
 
 ```

package/skills/aif-loop/SKILL.md

@@ -93,6 +93,26 @@ Read these files if present:
 
 Use them to keep outputs aligned with project conventions.
 
+**Read `.ai-factory/skill-context/aif-loop/SKILL.md`** — MANDATORY if the file exists.
+
+This file contains project-specific rules accumulated by `/aif-evolve` from patches,
+codebase conventions, and tech-stack analysis. These rules are tailored to the current project.
+
+**How to apply skill-context rules:**
+- Treat them as **project-level overrides** for this skill's general instructions
+- When a skill-context rule conflicts with a general rule written in this SKILL.md,
+  **the skill-context rule wins** (more specific context takes priority — same principle as nested CLAUDE.md files)
+- When there is no conflict, apply both: general rules from SKILL.md + project rules from skill-context
+- Do NOT ignore skill-context rules even if they seem to contradict this skill's defaults —
+  they exist because the project's experience proved the default insufficient
+- **CRITICAL:** skill-context rules apply to ALL outputs of this skill — including the generated
+  artifact, run state, and evaluation criteria. If a skill-context rule says "artifact MUST include X"
+  or "evaluation MUST check Y" — you MUST comply. Producing loop outputs that violate skill-context
+  rules is a bug.
+
+**Enforcement:** After generating any output artifact, verify it against all skill-context rules.
+If any rule is violated — fix the output before presenting it to the user.
+
 ## Step 0.1: Handle Non-Iteration Commands
 
 If command is `status`, `stop`, `list`, `history`, or `clean`, execute and stop: