@vheins/local-memory-mcp 0.14.7 → 0.14.8

package/dist/prompts/memory-index-policy.md

@@ -3,32 +3,26 @@ name: memory-index-policy
 description: Strict memory storage criteria.
 arguments: []
 agent: Memory Auditor
-version: "1.0.0"
-license: Proprietary — Personal Use Only
 category: workflows
-type: Utility
-complexity: Beginner
+version: "1.0.0"
 tags: [memory, indexing, policy, mcp]
-author: vheins
 ---
-# Memory Indexing Rules
 
-## ❌ FORBIDDEN
-- Temporary discussions/brainstorming.
-- Opinions without consensus.
-- Generic knowledge from public docs.
+## FSM
+
+Entry=G0 → S0 → S1 → S2  Exit=stored|rejected
+Guard: S(N) req S(N-1)✅
+
+G0 | is durable + project-specific? NOT forbidden types? | content provided? | → S0 / reject | —
+S0 | classify type (code_fact|decision|mistake|pattern|task_archive) + tech tags | G0✅ | classified | —
+S1 | scope: is_global ONLY if cross-repo applicable | S0✅ | scoped | —
+S2 | store via memory-store | S1✅ | memory created | —
 
-## ✅ MANDATORY
-Only store durable, project-specific knowledge.
-- **Types**: `code_fact`, `decision`, `mistake`, `pattern`, `task_archive`.
-- **Content**: Architecture, UI/UX choices, stack patterns, hard-won bug fixes.
-- **Global**: Set `is_global` only if applicable across repositories (e.g., framework anti-patterns).
-- **Categorization**: Use accurate technology tags.
+## Forbidden (G0→reject)
 
-## Operational Note
-- Do **not** store agent coordination state as memory.
-- Use `handoff-create`, `handoff-list`, and `handoff-update` for agent handoffs.
-- Use `task-claim` for task ownership instead of encoding claims in memory metadata.
-- File ownership/claims are coordination state; never store them as `file_claim` memory entries.
-- Use `standard-store` for normative coding standards; do not bury implementation rules in generic `decision` memories.
-- Use `standard-search` as the standards navigation layer before applying or creating standards.
+- Temporary discussions / brainstorming
+- Opinions without consensus
+- Generic knowledge from public docs
+- Agent coordination state (use handoff-create/handoff-list/handoff-update instead)
+- File ownership / claims (use task-claim/claim-list/claim-release instead)
+- Implementation rules (use standard-store instead)

package/dist/prompts/project-briefing.md

@@ -3,22 +3,20 @@ name: project-briefing
 description: Contextual onboarding to current repository.
 arguments: []
 agent: Session Concierge
-version: "1.0.0"
-license: Proprietary — Personal Use Only
 category: workflows
-type: Utility
-complexity: Beginner
+version: "1.0.0"
 tags: [workflow, briefing, onboarding, memory, backlog]
-author: vheins
 ---
-Initialize session in repository.
 
-Briefing Steps:
-1. **Repository**: Identify current repo from context.
-2. **Tasks**: Call `task-list` for `in_progress,pending` tasks.
-3. **Handoffs**: Call `handoff-list` with `status=pending` to surface transfer context. Treat only handoffs with unfinished work, blockers, next owner, or linked task as active.
-4. **Memory**: Call `memory-search` or `memory-recap` for recent decisions, patterns, and mistakes; hydrate important entries with `memory-detail`.
-5. **Standards**: Call `standard-search` with current repo/stack as the mandatory pre-implementation standards check. If no relevant standards are returned, say no applicable standards were found.
-6. **Core Context**: Summarize active task, pending handoffs, applicable standards, and top architectural decisions.
-7. **Priority Reminder**: Treat task priority with MCP semantics: `1=Low`, `2=Normal`, `3=Medium`, `4=High`, `5=Critical`.
-8. **Action**: Propose next steps based on the active queue.
+## FSM
+
+Entry=S0 → S1-4(parallel) → S5 → S6  Exit=briefed
+Guard: S(N) req S(N-1)✅
+
+S0 | identify current repo | — | repo context | —
+S1 | load tasks: task-list (in_progress, pending) | S0✅ | active tasks | —
+S2 | load handoffs: handoff-list (pending) — active only if unfinished work/blockers/linked task | S0✅ | pending transfers | —
+S3 | load memory: memory-search or memory-recap; hydrate via memory-detail | S0✅ | decisions, patterns, mistakes | —
+S4 | load standards: standard-search (repo, stack) | S0✅ | applicable standards | —
+S5 | summarize core: active task + pending handoffs + standards + top decisions | S1-4✅ | briefing | —
+S6 | propose next steps based on active queue; priority: 1=Low 2=Normal 3=Medium 4=High 5=Critical | S5✅ | action plan | —

package/dist/prompts/review-and-audit.md

@@ -6,55 +6,41 @@ arguments:
     description: Module, feature, or component to audit.
     required: false
 agent: Quality Auditor
-version: "1.0.0"
-license: Proprietary — Personal Use Only
 category: workflows
-type: Orchestrator
-complexity: Advanced
+version: "1.0.0"
 tags: [workflow, audit, ux, gap-analysis, mcp]
-author: vheins
 ---
-# Skill: review-and-audit (Audit Agent)
-
-## 1. ANALYSIS
-1. **Sequential Discovery**: Explore docs and code sequentially. NO parallel sub-agents.
-2. **UX Audit**: Use `chrome-dev-tools` for visual, navigation, and responsiveness checks.
-3. **Reference Audit**: Check current tool/prompt/resource definitions through code or the dashboard Reference flow.
-4. **Compare**: Match docs + code findings against live UI to find gaps/misalignments.
-
-## 🚫 FORBIDDEN: NON-EXECUTION
-DO NOT edit/create/delete files, run commands, or implement code.
-**Allowed**: Read code, `chrome-dev-tools`, `task-create`, `memory-store`, `task-list`, `memory-search`, `standard-search`, `handoff-list`, `handoff-update`.
-
-## ✅ OUTPUT: MCP ONLY
-ONLY call MCP tools. No prose, code, or external plans.
-
-## 2. PRE-TASK ANALYSIS
-1. **Search**: Call `memory-search` (Hybrid Search). 0.55 similarity threshold.
-2. **Standards**: Call `standard-search` before creating implementation tasks so task scope reflects applicable coding standards. If no relevant standards are returned, note that no applicable standards were found.
-3. **Handoffs**: Call `handoff-list` for pending transfer context related to the target. Treat handoffs as active only when they contain unfinished work, blockers, a next owner, or a linked task.
-4. **De-duplicate**: Call `task-list`. Skip existing/redundant tasks. Link via `parent_id`/`depends_on`.
-
-## 3. TASK DESIGN & FORMAT
-- **Atomic**: One change per task.
-- **Attributes**: `task_code`, `phase`, `priority`, `agent`, `model`.
-- **Priority Scale**: Use MCP ordering exactly: `1=Low`, `2=Normal`, `3=Medium`, `4=High`, `5=Critical`. `5` is the highest urgency.
-- **Description** (STRICT FORMAT):
-  ### 1. Context & Analysis
-  - **Finding**: Gap trigger.
-  - **Observation**: Reasoning.
-  - **Goal**: Clear objective.
-  ### 2. Target Files & Implementation
-  - Combined scope/steps per path/layer.
-  ### 3. Acceptance & Verification
-  - **Checklist**: `[ ]` criteria.
-  - **Testing**: Scenarios.
-
-## 4. LOGGING & MULTI-TASK
-- Log architectural/feature changes as `type: decision` via `memory-store`.
-- Log reusable coding rules as `standard-store` only when the rule is durable and source-backed.
-- Use Parent/Child structure for complex gaps.
-
-## 5. SELF-CHECK
-- ❌ No implementation.
-- ✅ ONLY MCP tool calls.
+
+## FSM
+
+Entry=S0 → S1 → S2 → S3  Exit=done
+Guard: S(N) req S(N-1)✅; NO code/edit/delete — read+MCP tools ONLY
+
+S0 | sequential discovery: docs → code → UI (chrome-dev-tools) | — | findings | —
+S1 | pre-task analysis: memory-search (0.55 threshold) + standard-search + handoff-list + task-list dedup | S0✅ | context | —
+S2 | design tasks: atomic, attributes (task_code, phase, priority, agent, model), strict description format | S1✅ | task specs | —
+S3 | create via task-create + log decisions via memory-store + standard-store for coding rules | S2✅ | MCP tasks | —
+
+## Description Format (STRICT — used in S2)
+
+```
+### 1. Context & Analysis
+- **Trigger**: Instruction/finding.
+- **Observation**: Technical reasoning.
+- **Goal**: Clear objective.
+### 2. Step & Implementation
+- Detailed execution steps per path/layer.
+### 3. Acceptance & Verification
+- **Checklist**: `[ ]` criteria.
+- **Testing**: Scenarios.
+```
+
+## Priority Scale
+
+`1=Low` `2=Normal` `3=Medium` `4=High` `5=Critical`
+
+## Logging
+
+- Decisions → `memory-store` (type=decision)
+- Reusable coding rules → `standard-store` (only if durable + source-backed)
+- Complex gaps → parent/child structure

package/dist/prompts/review-and-post-issue.md

@@ -12,43 +12,31 @@ arguments:
     description: Module, feature, or component to audit.
     required: false
 agent: Quality Auditor
-version: "1.0.0"
 category: workflows
+version: "1.0.0"
 tags: [workflow, audit, github, issue-triage]
 ---
 
-# Skill: review-and-post-issue (Audit Agent)
-
-## 1. ANALYSIS
-1. **Sequential Discovery**: Explore docs and code sequentially. NO parallel sub-agents.
-2. **UX Audit**: If applicable, use `chrome-dev-tools` for visual, navigation, and responsiveness checks.
-3. **Compare**: Match findings against live UI to find gaps/misalignments.
-
-## 🚫 FORBIDDEN: NON-EXECUTION
-DO NOT edit/create/delete files, run commands, or implement code.
-**Allowed**: Read code, `chrome-dev-tools`, `memory-search`, GitHub `search_issues`, `issue_write`.
+## FSM
 
-## ✅ OUTPUT: GITHUB ONLY
-ONLY call: `search_issues`, `issue_write` (method: 'create'), `memory-search`.
-No prose. No external plans.
+Entry=S0 → S1 → S2 → S3 Exit=done
+Guard: S(N) req S(N-1)✅; NO code/edit/delete — GitHub+MCP tools ONLY
 
-## 2. PRE-ISSUE ANALYSIS
-1. **Search**: Call `memory-search` (Hybrid Search). 0.55 similarity threshold.
-2. **De-duplicate**: Call `search_issues`. Skip existing/redundant issues. Comment on related issues if distinct.
+S0 | sequential discovery: docs → code → UI (chrome-dev-tools if applicable) | — | findings | —
+S1 | pre-issue analysis: memory-search (0.55 threshold) + search_issues dedup (comment on related if distinct) | S0✅ | context | —
+S2 | design issues: atomic, strict body format, labels | S1✅ | issue specs | —
+S3 | create via issue_write(method=create) | S2✅ | GitHub issues | —
 
-## 3. ISSUE DESIGN & FORMAT
-- **Atomic**: One change per issue.
-- **Body** (STRICT FORMAT):
-  ### 1. Context & Analysis
-  - **Finding**: Gap trigger.
-  - **Observation**: Reasoning.
-  - **Goal**: Clear objective.
-  ### 2. Target Files & Implementation
-  - Path/layer specific changes.
-  ### 3. Acceptance & Verification
-  - **Checklist**: `[ ]` criteria.
-  - **Testing**: Scenarios.
+## Issue Body Format (STRICT — used in S2)
 
-## 4. SELF-CHECK
-- ❌ No implementation.
-- ✅ ONLY GitHub/Memory tool calls.
+```
+### 1. Context & Analysis
+- **Trigger**: Instruction/finding.
+- **Observation**: Technical reasoning.
+- **Goal**: Clear objective.
+### 2. Step & Implementation
+- Detailed execution steps per path/layer.
+### 3. Acceptance & Verification
+- **Checklist**: `[ ]` criteria.
+- **Testing**: Scenarios.
+```

package/dist/prompts/root-cause-analysis.md

@@ -12,20 +12,19 @@ arguments:
     description: Logs, errors, metrics.
     required: false
 agent: Diagnostic Lead
-version: "1.0.0"
 category: debugging
+version: "1.0.0"
 tags: [root-cause, 5-why, debugging, diagnosis]
 ---
 
-Conduct root cause analysis for repository bug.
+## FSM
+
+Entry=S0 → S1 → S2 → S3  Exit=diagnosis
+Guard: S(N) req S(N-1)✅
 
-Stack: {{tech_stack}}
-Bug: {{bug_description}}
-Symptoms: {{symptoms}}
+S0 | restate symptom: technical problem statement | tech_stack + bug_description provided? | symptom statement | —
+S1 | 5-why analysis: causal chain from symptom to core failure | S0✅ | causal chain | —
+S2 | identify root cause: "root cause is [X] because [Y], allowing [Z]" | S1✅ | root cause | —
+S3 | recommend fix addressing root cause + prevention (monitoring/test) | S2✅ | recommendation | —
 
-Output:
-1. **Symptom**: Technical problem restatement.
-2. **5-Whys**: Causal chain from symptom to core failure.
-3. **Root Cause**: "The root cause is [X] because [Y], allowing [Z]."
-4. **Recommendation**: Fix addressing root cause.
-5. **Prevention**: Monitoring/testing measure.
+Stack: {{tech_stack}} Bug: {{bug_description}} Symptoms: {{symptoms}}

package/dist/prompts/security-triage.md

@@ -12,20 +12,20 @@ arguments:
     description: Usage context.
     required: false
 agent: Security Engineer
-version: "1.0.0"
 category: debugging
+version: "1.0.0"
 tags: [security, triage, vulnerability, cvss, appsec]
 ---
 
-Triage vulnerability for repository.
+## FSM
+
+Entry=S0 → S1 → S2 → S3 → S4  Exit=assessment
+Guard: S(N) req S(N-1)✅
 
-Stack: {{tech_stack}}
-Report: {{vulnerability_report}}
-Context: {{codebase_context}}
+S0 | classify: type, CVE, CVSS vector, score | tech_stack + vuln_report provided? | classification | —
+S1 | assess exploitability: reachability + attack scenarios | S0✅ | exploit scenarios | —
+S2 | assess impact: CIA triad | S1✅ | impact assessment | —
+S3 | remediate: priority P0-P3 + fix steps | S2✅ | remediation plan | —
+S4 | verify: testing method to confirm fix | S3✅ | verification plan | —
 
-Output:
-1. **Classification**: Type, CVE, CVSS, vector.
-2. **Exploitability**: Reachability & scenarios.
-3. **Impact**: CIA impact.
-4. **Remediation**: Priority (P0-P3) & fix steps.
-5. **Verification**: Testing method.
+Stack: {{tech_stack}} Report: {{vulnerability_report}} Context: {{codebase_context}}

package/dist/prompts/senior-code-review.md

@@ -9,28 +9,19 @@ arguments:
     description: Production context (SLA, data, conventions).
     required: false
 agent: Principal Reviewer
-version: "1.1.0"
+version: "1.2.0"
 category: coding
 tags: [code-review, production-readiness, security, observability, senior-review, architecture]
 ---
 
-Perform production-readiness review for repository.
+## FSM
 
-Stack: {{tech_stack}}
-Context: {{context}}
+Entry=S0 → S1 → S2 → S3  Exit=decision
+Guard: S(N) req S(N-1)✅; cite code evidence only; ONE fix option per finding
 
-Audit Dimensions:
-1. **Errors**: Completeness & patterns.
-2. **Security**: Validation, injection, secrets.
-3. **Performance**: Complexity, DB efficiency.
-4. **Observability**: Logs, metrics, traces.
-5. **Testing**: Coverage & quality.
-6. **Docs**: Clarity & accuracy.
+S0 | audit 6 dimensions: errors, security, performance (N+1, cache, complexity), observability (logs, metrics, traces), testing (coverage, quality), docs (clarity) | tech_stack provided? | findings[] | —
+S1 | check cross-domain invariants: lifecycle, concurrency guard, derived state, upload safety, file>500→refactor, doc hierarchy | S0✅ | invariant results | —
+S2 | assign severity: CRITICAL (bug/data loss) | HIGH (concurrency/arch) | MEDIUM (maintainability) | LOW (cosmetic) | S0-1✅ | scored findings | —
+S3 | produce: DECISION (APPROVE|REQUEST_CHANGES|NOT_READY) + SEVERITY_SCORE + MESSAGE (blockers only) | S2✅ | review decision | —
 
-Output per Finding:
-- **Severity**: P0-P3.
-- **Problem**: What & why.
-- **Location**: Path/function.
-- **Fix**: Actionable step.
-
-Verdict: READY | READY WITH MINOR FIXES | NOT READY
+Stack: {{tech_stack}} Context: {{context}}

package/dist/prompts/sentinel-issue-resolver.md

@@ -6,46 +6,24 @@ arguments:
     description: The full URL of the GitHub issue to resolve.
     required: true
 agent: SENTINEL Issue Resolver
-version: "1.0.0"
-license: Proprietary — Personal Use Only
 category: workflows
-type: Orchestrator
-complexity: Advanced
+version: "1.0.0"
 tags: [workflow, github, issue-resolution, sentinel]
-author: vheins
 ---
 
-# SENTINEL Protocol
-
-You are **SENTINEL**, an elite issue resolution agent. Your primary objective is to eliminate errors and fulfill requirements described in GitHub issues with surgical precision.
-
-## 1. INTELLIGENCE GATHERING
-1. **Analyze Issue**: Use `issue_read` to fetch the main description AND all comments. Comments often contain critical root cause analysis, reproduction steps, or specific user requirements.
-2. **Context Synthesis**: Combine the issue data with local codebase knowledge. Search project memory (`memory-search`) and coding standards (`standard-search`) to ensure your fix aligns with existing architecture.
-3. **Task Registration**: Use `task-create` to register your plan in MCP. Link the task to the GitHub Issue URL in the metadata.
-
-## 2. EXECUTION & RESOLUTION
-1. **Claim Work**: Use `task-claim` for the generated task.
-2. **Implement Fix**: Perform the necessary code changes. Ensure all changes follow established project conventions and pass existing tests.
-3. **Validate**: Run tests and linters to verify the fix. Perform end-to-end verification if applicable.
-
-## 3. FINALIZATION & COMMIT
-1. **Identity**: Use the local Git configuration (name/email) for all commits.
-2. **Commit Format**: Every commit MUST follow this specific structure:
-   ```
-   type(scope): your commit message
-
-   - {{task_title}}
-     {{summary_task}}
-
-   {{keyword}} #{{issue_number}}
-   ```
-   Use `fix` for bug fixes, `closes` for features/chores, `resolve` as general. Extract the issue number from the provided `issue_url`.
+## FSM
 
-3. **MCP Update**: Transition the task to `completed` with a detailed comment linking to the resolution.
-4. **Issue Closure**: If authorized or part of the workflow, add a final comment to the GitHub issue summarizing the fix.
+Entry=S0 → S1 → S2 → S3 → S4 → S5 → S6 → S7 → S8  Exit=resolved
+Guard: S(N) req S(N-1)✅; autonomous — no permission per step
 
-## ✅ OUTPUT: AUTONOMOUS ACTION
-Do not ask for permission for each step. Analyze, plan, fix, and verify. Provide a final summary of the resolution to the user.
+S0 | fetch: issue_read body + ALL comments (get_comments) | issue_url provided? | raw issue + all comments | —
+S1 | analyze comments: extract requirements, hints, root cause clues, reproduction steps, error details | S0✅ | comment analysis | —
+S2 | detect attachments: parse all comments for image/video markdown URLs, linked assets | S1✅ | attachment URL list | —
+S3 | download: gh CLI (gh issue view, gh api) to fetch each attachment — private repo, NO curl | S2✅ & attachments exist? | local files | —
+S4 | analyze attachments: inspect screenshots/videos for UI bugs, error states, configs, visual hints | S3✅ | attachment analysis | —
+S5 | research: memory-search + standard-search + codebase exploration (trace call sites, read docs) | S0-4✅ | full context | —
+S6 | register: task-create (link issue URL) + task-claim + task-update→in_progress | S5✅ | MCP task | —
+S7 | implement fix + validate: tests, linters, e2e | S6✅ | verified changes | —
+S8 | finalize: commit (type(scope): msg — fix #N) + task-update→completed + issue comment summary | S7✅ | resolution | —
 
-Target Issue: {{issue_url}}
+Target: {{issue_url}}

package/dist/prompts/session-planner.md

@@ -6,21 +6,21 @@ arguments:
     description: High-level session goal.
     required: true
 agent: Strategy Lead
-version: "1.0.0"
 category: workflows
+version: "1.0.0"
 tags: [workflow, planning, task-breakdown]
 ---
 
-Plan execution for: '{{objective}}'.
+## FSM
+
+Entry=S0 → S1 → S2 → S3 → S4 → S5  Exit=planned
+Guard: S(N) req S(N-1)✅
 
-Steps:
-1. **Orient**: Call `task-list` to avoid duplicate active/backlog work.
-2. **Standards**: Call `standard-search` for objectives that may lead to code edits, test edits, refactors, migrations, or implementation decisions. If no relevant standards are returned, state that no applicable standards were found.
-3. **Handoffs**: Call `handoff-list` for pending context that may affect sequencing. Stale pending handoffs that only summarize completed work should be closed with `handoff-update`, not planned as queue work.
-4. **Analyze**: Break into 3-7 atomic, verifiable tasks.
-5. **Phase**: Group into `research`, `implementation`, and `validation`.
-6. **Hierarchy**: Use `parent_id` / `depends_on` for sequencing.
-7. **Priority Scale**: When creating tasks, use the exact MCP scale `1=Low`, `2=Normal`, `3=Medium`, `4=High`, `5=Critical`. Higher number means higher urgency.
-8. **Create**: Use `task-create` in current repo with stable `task_code`, tags, and acceptance criteria.
+S0 | orient: task-list (avoid dupes) + standard-search (if code edits) + handoff-list (close stale) | objective provided? | existing state | —
+S1 | analyze: break into 3-7 atomic verifiable tasks | S0✅ | task breakdown | —
+S2 | phase: group into research / implementation / validation | S1✅ | phased tasks | —
+S3 | hierarchy: parent_id + depends_on for sequencing; priority 1-5 scale | S2✅ | structured plan | —
+S4 | create via task-create (stable task_code, tags, acceptance criteria) | S3✅ | MCP tasks | —
+S5 | display final plan to user | S4✅ | user output | —
 
-Display final plan to user.
+Objective: {{objective}}

package/dist/prompts/task-management-guidelines.md

@@ -3,36 +3,27 @@ name: task-management-guidelines
 description: Task tracking & progress management standards.
 arguments: []
 agent: Project Manager
-version: "1.0.0"
-license: Proprietary — Personal Use Only
 category: workflows
-type: Utility
-complexity: Beginner
+version: "1.0.0"
 tags: [workflow, tasks, status-management, mcp]
-author: vheins
 ---
-# Task Management Standards
 
-## 1. NAVIGATION (`task-list`)
--   **Sync**: Call `task-list` at every session start (default: `in_progress,pending`).
--   **Format**: Compact pointer table: `id`, `task_code`, `title`, `status`, `priority`, `updated_at`, `comments_count`. Use `query` for keyword search.
--   **Priority Scale**: Interpret `priority` with MCP semantics: `1=Low`, `2=Normal`, `3=Medium`, `4=High`, `5=Critical`.
--   **Retrieve**: Fetch full context via `task-detail` AFTER selecting a task. The hydrated task includes current coordination state such as active claims and pending handoffs.
--   **Coordination**: Check active ownership with task coordination metadata, `task-claim`, and `claim-list`. NEVER work on tasks claimed by others. Focus on ONE task at a time.
+## FSM — Task Lifecycle
+
+Entry=S0 → S1 → S2 → S3 → S4   Exit=done|archived
+Guard: S(N) req S(N-1)✅; MUST transition backlog/pending → in_progress → completed
+
+S0 | plan: task-create for full lifecycle (Research→Strategy→Execution→Validation) | — | tasks | —
+S1 | claim: task-claim with task_code; inspect via claim-list | S0✅, not claimed by others | ownership | —
+S2 | progress: task-update → in_progress | S1✅ | active work | —
+S3 | validate: tests pass / explicit doc why verification skipped | S2✅ | evidence | —
+S4 | complete: task-update → completed (auto-releases claims, expires handoffs, archives) | S3✅ | completion | —
+
+## FSM — Navigation
 
-## 2. DETAIL TOOLS
--   **Tasks**: Call `task-detail` for history/comments (ID or `task_code`).
--   **Memory**: Call `memory-detail` for full entry content.
--   **Standards**: Call `standard-search` before any code edit, test edit, refactor, migration, or implementation decision. If no relevant standards are returned, continue and state that no applicable standards were found.
--   **Handoffs**: Call `handoff-list` to discover pending context transfers before starting a task. Close stale handoffs with `handoff-update` when no concrete next owner, unfinished task, or blocker remains.
+Entry=S0 → S1 → S2   Exit=mutated
+Guard: S(N) req S(N-1)✅
 
-## 3. WORKFLOW
--   **Planning**: Create tasks for full lifecycle (Research → Strategy → Execution → Validation).
--   **Transition Safety**: MUST move from `backlog/pending` → `in_progress` → `completed`. Skipping `in_progress` is forbidden.
--   **Automatic Cleanup**: `task-update` to `completed` or `canceled` automatically releases active claims and expires pending handoffs linked to that task.
--   **Claiming**: Use `task-claim` when taking ownership of a task, with `task_code` when working from human-visible queues.
--   **Claim Inspection**: Use `claim-list` when ownership is unclear or when triaging stale work.
--   **Claim Release**: Use `claim-release` to clear stale ownership explicitly when a task is being handed back or reassigned.
--   **Handoff**: Use `handoff-create` only when pausing or transferring unfinished work. Do not use pending handoffs for completion summaries; close consumed/stale handoffs with `handoff-update`.
--   **Validation**: Only mark `completed` after passing tests or explicitly documenting why verification could not run.
--   **Archiving**: Completion triggers auto-archive to `task_archive` memory with token reporting.
+S0 | list: task-list (in_progress, pending) | — | pointer table | —
+S1 | detail: task-detail after selecting (includes claims + handoffs) | S0✅ | full context | —
+S2 | mutate: task-create | task-update | task-delete | S1✅ | changed | —