@vheins/local-memory-mcp 0.5.13 → 0.5.15

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (17) hide show
  1. package/dist/dashboard/public/assets/{index-DXvlmarr.css → index-C8FB4maW.css} +1 -1
  2. package/dist/dashboard/public/assets/index-CUNSfwmD.js +78 -0
  3. package/dist/dashboard/public/index.html +2 -2
  4. package/dist/mcp/prompts/definitions/create-task.md +12 -12
  5. package/dist/mcp/prompts/definitions/documentation-sync.md +1 -1
  6. package/dist/mcp/prompts/definitions/import-github-issues.md +1 -1
  7. package/dist/mcp/prompts/definitions/learning-retrospective.md +1 -1
  8. package/dist/mcp/prompts/definitions/memory-agent-core.md +1 -1
  9. package/dist/mcp/prompts/definitions/memory-guided-review.md +1 -1
  10. package/dist/mcp/prompts/definitions/project-briefing.md +2 -2
  11. package/dist/mcp/prompts/definitions/review-and-audit.md +12 -12
  12. package/dist/mcp/prompts/definitions/review-and-post-issue.md +3 -3
  13. package/dist/mcp/prompts/definitions/task-management-guidelines.md +1 -1
  14. package/dist/mcp/prompts/definitions/task-memory-executor.md +6 -6
  15. package/dist/mcp/prompts/definitions/tech-affinity-scout.md +1 -1
  16. package/package.json +1 -1
  17. package/dist/dashboard/public/assets/index-B1H-cHq2.js +0 -78
package/dist/dashboard/public/index.html CHANGED
@@ -8,8 +8,8 @@
8
8
  <link rel="preconnect" href="https://fonts.googleapis.com">
9
9
  <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
10
10
  <link href="https://fonts.googleapis.com/css2?family=Inter:wght@300;400;500;600;700;800&family=JetBrains+Mono:wght@400;500&display=swap" rel="stylesheet">
11
- <script type="module" crossorigin src="/assets/index-B1H-cHq2.js"></script>
12
- <link rel="stylesheet" crossorigin href="/assets/index-DXvlmarr.css">
11
+ <script type="module" crossorigin src="/assets/index-CUNSfwmD.js"></script>
12
+ <link rel="stylesheet" crossorigin href="/assets/index-C8FB4maW.css">
13
13
  </head>
14
14
  <body>
15
15
  <div id="app"></div>
package/dist/mcp/prompts/definitions/create-task.md CHANGED
@@ -28,20 +28,20 @@ You are **STRICTLY FORBIDDEN** from performing any of the following actions:
28
28
 
29
29
  **Allowed Actions:**
30
30
  * Read code and analyze context
31
- * Create tasks via `@vheins/local-memory-mcp tools task-create`
32
- * Record decisions via `@vheins/local-memory-mcp tools memory-store`
33
- * List tasks via `@vheins/local-memory-mcp tools task-list`
34
- * Search memory via `@vheins/local-memory-mcp tools memory-search`
31
+ * Create tasks via `local-memory-mcp` MCP tools `task-create`
32
+ * Record decisions via `local-memory-mcp` MCP tools `memory-store`
33
+ * List tasks via `local-memory-mcp` MCP tools `task-list`
34
+ * Search memory via `local-memory-mcp` MCP tools `memory-search`
35
35
 
36
36
  ---
37
37
 
38
38
  ### ✅ ALLOWED OUTPUT (STRICT)
39
39
 
40
40
  Your output MUST ONLY consist of calls to:
41
- * `@vheins/local-memory-mcp tools task-create`
42
- * `@vheins/local-memory-mcp tools memory-store`
43
- * `@vheins/local-memory-mcp tools task-list`
44
- * `@vheins/local-memory-mcp tools memory-search`
41
+ * `local-memory-mcp` MCP tools `task-create`
42
+ * `local-memory-mcp` MCP tools `memory-store`
43
+ * `local-memory-mcp` MCP tools `task-list`
44
+ * `local-memory-mcp` MCP tools `memory-search`
45
45
 
46
46
  **❌ DO NOT:**
47
47
  * Output explanations or narrative text
@@ -53,8 +53,8 @@ Your output MUST ONLY consist of calls to:
53
53
  ### 1. PRE-ANALYSIS (MANDATORY)
54
54
 
55
55
  Before creating tasks, you MUST:
56
- 1. **Context discovery**: Call `@vheins/local-memory-mcp tools memory-search` to query existing architectural and historical context.
57
- 2. **Sync backlog**: Call `@vheins/local-memory-mcp tools task-list` to check existing tasks. **CRITICAL: Do NOT create a new task if a similar, redundant task already exists in `backlog` or `pending` status. If your new findings are distinct but related to an existing task, link them using `parent_id` or `depends_on` instead of creating an isolated task.**
56
+ 1. **Context discovery**: Call `local-memory-mcp` MCP tools `memory-search` to query existing architectural and historical context.
57
+ 2. **Sync backlog**: Call `local-memory-mcp` MCP tools `task-list` to check existing tasks. **CRITICAL: Do NOT create a new task if a similar, redundant task already exists in `backlog` or `pending` status. If your new findings are distinct but related to an existing task, link them using `parent_id` or `depends_on` instead of creating an isolated task.**
58
58
 
59
59
  ---
60
60
 
@@ -71,7 +71,7 @@ Each task MUST be:
71
71
 
72
72
  ### 3. TASK ATTRIBUTES (MANDATORY)
73
73
 
74
- Each `@vheins/local-memory-mcp tools task-create` MUST include:
74
+ Each `local-memory-mcp` MCP tools `task-create` MUST include:
75
75
  * `task_code`: (e.g., FEAT-123 / FIX-456 / REFACTOR-789)
76
76
  * `phase`: (Discovery / Implementation / Testing)
77
77
  * `priority`: (1–5)
@@ -103,7 +103,7 @@ The `description` field MUST follow this structure EXACTLY:
103
103
  ---
104
104
 
105
105
  ### 4. MEMORY STORAGE (CONDITIONAL)
106
- If the instruction or prompt involves a decision, new feature, or architectural change, you MUST log it as a memory using `@vheins/local-memory-mcp tools memory-store` with `type: decision`. This ensures the decision to create the task and its triggering context is captured in the global memory.
106
+ If the instruction or prompt involves a decision, new feature, or architectural change, you MUST log it as a memory using `local-memory-mcp` MCP tools `memory-store` with `type: decision`. This ensures the decision to create the task and its triggering context is captured in the global memory.
107
107
  **CRITICAL**: Do NOT log tasks as decisions if they are purely for bug fixes or straightforward defect resolutions.
108
108
 
109
109
  ---
package/dist/mcp/prompts/definitions/documentation-sync.md CHANGED
@@ -7,7 +7,7 @@ agent: Documentation Specialist
7
7
  Please verify if our local documentation (README.md, docs/*.md, .agents/documents/**/*.md, .kiro/**/*.md) is in sync with our stored memories for the current repository.
8
8
 
9
9
  Steps:
10
- 1. **Fetch Decisions**: Use '@vheins/local-memory-mcp tools memory-search' to find all 'decision' type memories for this repo.
10
+ 1. **Fetch Decisions**: Use `local-memory-mcp` MCP tools `memory-search` to find all 'decision' type memories for this repo.
11
11
  2. **Read Docs**: Read the primary project documentation files including those in .agents/documents and .kiro.
12
12
  3. **Identify Gaps**: Is there any durable knowledge in the memory that is MISSING from the docs? Is there any documentation that is OUTDATED based on recent decisions?
13
13
  4. **Propose Updates**: Suggest specific changes to the documentation to reflect the current source of truth.
package/dist/mcp/prompts/definitions/import-github-issues.md CHANGED
@@ -10,7 +10,7 @@ Please follow these steps:
10
10
 
11
11
  1. **Access Issues**: You MUST use the `github-mcp-server` integration to fetch open issues for the current repository.
12
12
  * **Fallback**: If the GitHub MCP integration is unavailable or throws an error, fallback to using the GitHub CLI via terminal (`gh issue list --json number,title,body,labels,url`).
13
- 2. **Review Existing Tasks**: Call '@vheins/local-memory-mcp tools task-list' for the current repository to identify tasks already imported.
13
+ 2. **Review Existing Tasks**: Call `local-memory-mcp` MCP tools `task-list` for the current repository to identify tasks already imported.
14
14
  3. **Map and Create**: For each relevant issue that hasn't been imported yet:
15
15
  - Use 'task-manage' with action='create'.
16
16
  - Set 'task_code' to 'GH-{{issue_number}}' (e.g., GH-123).
package/dist/mcp/prompts/definitions/learning-retrospective.md CHANGED
@@ -14,4 +14,4 @@ Please reflect on the changes and identify knowledge worth keeping:
14
14
  2. **Decisions**: Did we make a choice between multiple options (e.g., library choice, UI pattern)? (Store as 'decision')
15
15
  3. **Patterns**: Did we establish a repeatable way of doing things in this codebase? (Store as 'pattern')
16
16
 
17
- Use '@vheins/local-memory-mcp tools memory-store' to record any high-value findings. Be concise and use appropriate technology tags.
17
+ Use `local-memory-mcp` MCP tools `memory-store` to record any high-value findings. Be concise and use appropriate technology tags.
package/dist/mcp/prompts/definitions/memory-agent-core.md CHANGED
@@ -18,7 +18,7 @@ Core Behavioral Rules:
18
18
  2. Never repeat known mistakes documented in memory.
19
19
  3. Never use memory from another repository UNLESS it shares the same technology tags (Affinity) or is marked as Global.
20
20
  4. If memory conflicts with the user's new request, detect the conflict and ask for clarification or propose a 'supersedes' update.
21
- 5. After using a memory to generate code, you MUST call '@vheins/local-memory-mcp tools memory-acknowledge' to report its utility.
21
+ 5. After using a memory to generate code, you MUST call `local-memory-mcp` MCP tools `memory-acknowledge` to report its utility.
22
22
 
23
23
  Memory Usage Policy:
24
24
  Before generating code:
package/dist/mcp/prompts/definitions/memory-guided-review.md CHANGED
@@ -10,6 +10,6 @@ agent: Code Auditor
10
10
  Please review the code in '{{file_path}}'.
11
11
 
12
12
  Your goal is to ensure compliance with our stored project knowledge for the current repository:
13
- 1. **Search Constraints**: Use '@vheins/local-memory-mcp tools memory-search' with current_file_path='{{file_path}}' and the current repo context to find relevant decisions and patterns.
13
+ 1. **Search Constraints**: Use `local-memory-mcp` MCP tools `memory-search` with current_file_path='{{file_path}}' and the current repo context to find relevant decisions and patterns.
14
14
  2. **Evaluate Compliance**: Does the code follow established patterns? Does it repeat any known mistakes?
15
15
  3. **Feedback**: Provide specific suggestions for improvement if you find violations of stored decisions.
package/dist/mcp/prompts/definitions/project-briefing.md CHANGED
@@ -7,7 +7,7 @@ agent: Session Concierge
7
7
  I am starting a new session in the current repository.
8
8
 
9
9
  Please perform a briefing to catch up on the project:
10
- 1. **Recent Knowledge**: Call '@vheins/local-memory-mcp tools memory-search' with a broad query or relevant tags for the current repo to discover key decisions, patterns, and mistakes without the noise of a full recap.
11
- 2. **Current Tasks**: Call '@vheins/local-memory-mcp tools task-list' to understand what is currently pending or in-progress.
10
+ 1. **Recent Knowledge**: Call `local-memory-mcp` MCP tools `memory-search` with a broad query or relevant tags for the current repo to discover key decisions, patterns, and mistakes without the noise of a full recap.
11
+ 2. **Current Tasks**: Call `local-memory-mcp` MCP tools `task-list` to understand what is currently pending or in-progress.
12
12
  3. **Context Check**: Summarize the top 3 most important architectural decisions you found.
13
13
  4. **Readiness**: Tell me what you are ready to help with based on the current backlog.
package/dist/mcp/prompts/definitions/review-and-audit.md CHANGED
@@ -37,19 +37,19 @@ You are **STRICTLY FORBIDDEN** from performing any of the following actions:
37
37
  **Allowed Actions:**
38
38
  * Read code and analyze context
39
39
  * Use `chrome-dev-tools` MCP to inspect browser UX
40
- * Create tasks via `@vheins/local-memory-mcp tools task-create`
41
- * Record decisions via `@vheins/local-memory-mcp tools memory-store`
42
- * List tasks via `@vheins/local-memory-mcp tools task-list`
43
- * Search memory via `@vheins/local-memory-mcp tools memory-search`
40
+ * Create tasks via `local-memory-mcp` MCP tools `task-create`
41
+ * Record decisions via `local-memory-mcp` MCP tools `memory-store`
42
+ * List tasks via `local-memory-mcp` MCP tools `task-list`
43
+ * Search memory via `local-memory-mcp` MCP tools `memory-search`
44
44
 
45
45
  ---
46
46
 
47
47
  ### ✅ ALLOWED OUTPUT (STRICT)
48
48
  If gaps are found, your output MUST ONLY consist of calls to:
49
- * `@vheins/local-memory-mcp tools task-create`
50
- * `@vheins/local-memory-mcp tools memory-store`
51
- * `@vheins/local-memory-mcp tools task-list`
52
- * `@vheins/local-memory-mcp tools memory-search`
49
+ * `local-memory-mcp` MCP tools `task-create`
50
+ * `local-memory-mcp` MCP tools `memory-store`
51
+ * `local-memory-mcp` MCP tools `task-list`
52
+ * `local-memory-mcp` MCP tools `memory-search`
53
53
 
54
54
  **❌ DO NOT:**
55
55
  * Output explanations or narrative text
@@ -60,8 +60,8 @@ If gaps are found, your output MUST ONLY consist of calls to:
60
60
 
61
61
  ### 3. PRE-ANALYSIS FOR TASK GENERATION (MANDATORY)
62
62
  Before creating tasks, you MUST:
63
- 1. **Context discovery**: Call `@vheins/local-memory-mcp tools memory-search` to query existing architectural and historical context.
64
- 2. **Sync backlog**: Call `@vheins/local-memory-mcp tools task-list` to check existing tasks. **CRITICAL: Do NOT create a new task if a similar, redundant task already exists in `backlog` or `pending` status. If your new findings are distinct but related to an existing task, link them using `parent_id` or `depends_on` instead of creating an isolated task.**
63
+ 1. **Context discovery**: Call `local-memory-mcp` MCP tools `memory-search` to query existing architectural and historical context.
64
+ 2. **Sync backlog**: Call `local-memory-mcp` MCP tools `task-list` to check existing tasks. **CRITICAL: Do NOT create a new task if a similar, redundant task already exists in `backlog` or `pending` status. If your new findings are distinct but related to an existing task, link them using `parent_id` or `depends_on` instead of creating an isolated task.**
65
65
 
66
66
  ---
67
67
 
@@ -76,7 +76,7 @@ Each task MUST be:
76
76
  ---
77
77
 
78
78
  ### 5. TASK ATTRIBUTES (MANDATORY)
79
- Each `@vheins/local-memory-mcp tools task-create` MUST include:
79
+ Each `local-memory-mcp` MCP tools `task-create` MUST include:
80
80
  * `task_code`: (e.g., FEAT-123 / FIX-456 / REFACTOR-789)
81
81
  * `phase`: (Discovery / Implementation / Testing)
82
82
  * `priority`: (1–5)
@@ -108,7 +108,7 @@ The `description` field MUST follow this structure EXACTLY:
108
108
  ---
109
109
 
110
110
  ### 6. MEMORY STORAGE (CONDITIONAL)
111
- If the finding or gap involves a decision, new feature, or architectural change, you MUST log it as a memory using `@vheins/local-memory-mcp tools memory-store` with `type: decision`. This ensures the decision to create the task and its triggering context is captured in the global memory.
111
+ If the finding or gap involves a decision, new feature, or architectural change, you MUST log it as a memory using `local-memory-mcp` MCP tools `memory-store` with `type: decision`. This ensures the decision to create the task and its triggering context is captured in the global memory.
112
112
  **CRITICAL**: Do NOT log tasks as decisions if they are purely for bug fixes or straightforward defect resolutions.
113
113
 
114
114
  ---
package/dist/mcp/prompts/definitions/review-and-post-issue.md CHANGED
@@ -42,7 +42,7 @@ You are **STRICTLY FORBIDDEN** from performing any of the following actions:
42
42
  **Allowed Actions:**
43
43
  * Read code and analyze context
44
44
  * Use `chrome-dev-tools` MCP to inspect browser UX
45
- * Search memory via `@vheins/local-memory-mcp tools memory-search`
45
+ * Search memory via `local-memory-mcp` MCP tools `memory-search`
46
46
  * Search GitHub issues via **Github MCP Server Tools (search_issues)**
47
47
  * Create GitHub issues via **Github MCP Server Tools (issue_write)** (method: 'create')
48
48
 
@@ -52,7 +52,7 @@ You are **STRICTLY FORBIDDEN** from performing any of the following actions:
52
52
  If gaps are found, your output MUST ONLY consist of calls to:
53
53
  * **Github MCP Server Tools (search_issues)**
54
54
  * **Github MCP Server Tools (issue_write)**
55
- * `@vheins/local-memory-mcp tools memory-search`
55
+ * `local-memory-mcp` MCP tools `memory-search`
56
56
 
57
57
  **❌ DO NOT:**
58
58
  * Output explanations or narrative text
@@ -63,7 +63,7 @@ If gaps are found, your output MUST ONLY consist of calls to:
63
63
 
64
64
  ### 3. PRE-ANALYSIS FOR ISSUE GENERATION (MANDATORY)
65
65
  Before creating issues, you MUST:
66
- 1. **Context discovery**: Call `@vheins/local-memory-mcp tools memory-search` to query existing architectural and historical context.
66
+ 1. **Context discovery**: Call `local-memory-mcp` MCP tools `memory-search` to query existing architectural and historical context.
67
67
  2. **Sync GitHub Backlog**: Call **Github MCP Server Tools (search_issues)** with relevant keywords to check for existing issues. **CRITICAL: Do NOT create a new issue if a similar, redundant issue already exists. If your findings are distinct but related, comment on the existing issue instead.**
68
68
 
69
69
  ---
package/dist/mcp/prompts/definitions/task-management-guidelines.md CHANGED
@@ -14,7 +14,7 @@ Guidelines for Task Management:
14
14
  - Use 'metadata' to store technical debt notes or implementation details.
15
15
 
16
16
  2. Resource & Tool Usage:
17
- - MANDATORY: Always call '@vheins/local-memory-mcp tools task-list' at the very start of a new session to understand current progress and avoid duplicating work.
17
+ - MANDATORY: Always call `local-memory-mcp` MCP tools `task-list` at the very start of a new session to understand current progress and avoid duplicating work.
18
18
  - Resource: You can also read 'tasks://current' for a filtered view of active tasks for the current repository.
19
19
  - Coordinate: If a task is already 'in_progress', do not attempt to work on it unless specifically asked to collaborate.
20
20
 
package/dist/mcp/prompts/definitions/task-memory-executor.md CHANGED
@@ -19,19 +19,19 @@ description: Execute all pending tasks for the current repository
19
19
  Please follow this strict execution flow:
20
20
 
21
21
  1. **Identify Repository**: Determine the current repository name (e.g., from git config or workspace context).
22
- 2. **Fetch Tasks**: Call '@vheins/local-memory-mcp tools task-list' for the identified repository for statuses 'pending' and 'in_progress'.
22
+ 2. **Fetch Tasks**: Call `local-memory-mcp` MCP tools `task-list` for the identified repository for statuses 'pending' and 'in_progress'.
23
23
  3. **Filter Stale**: Identify 'in_progress' tasks that are **stale** (stale is defined as > 30 Minutes without update, often because an agent stopped or crashed).
24
24
  4. **Single-Task Execution Loop (STRICT)**: You MUST process tasks EXACTLY ONE AT A TIME. DO NOT update multiple tasks to 'in_progress' simultaneously. For the SINGLE task you select:
25
- - **Start**: Call '@vheins/local-memory-mcp tools task-update' to set status='in_progress' for this task ONLY. Provide current agent/role information in the metadata.
25
+ - **Start**: Call `local-memory-mcp` MCP tools `task-update` to set status='in_progress' for this task ONLY. Provide current agent/role information in the metadata.
26
26
  - **Execute**: Perform the work described in the task title and description.
27
27
  - **Inspect Codebase Logic & Documentation First (MANDATORY)**: Before marking anything done, inspect the relevant code paths, call sites, configs, tests, and affected modules in the repository. Also read the relevant documentation, as it might need to be updated or fixed. Do not infer correctness from file presence alone.
28
28
  - **Validate Behavior (MANDATORY)**: Ensure the implementation logic satisfies the task intent and follows project standards. Validation must focus on behavior, control flow, data flow, and integration points, not just whether a file/class/function exists.
29
- - **Complete Only With Evidence**: Call '@vheins/local-memory-mcp tools task-update' to set status='completed' only after recording concrete evidence in the 'comment' field. The comment must include: files inspected, logic verified, checks/tests run (or why they could not run), and the exact reason the task is considered complete.
30
- - **Compact Context**: Summarize key learnings, decisions, and patterns discovered during task execution. Store critical insights as memory entries (type: 'code_fact' or 'pattern') using '@vheins/local-memory-mcp tools memory-store'.
29
+ - **Complete Only With Evidence**: Call `local-memory-mcp` MCP tools `task-update` to set status='completed' only after recording concrete evidence in the 'comment' field. The comment must include: files inspected, logic verified, checks/tests run (or why they could not run), and the exact reason the task is considered complete.
30
+ - **Compact Context**: Summarize key learnings, decisions, and patterns discovered during task execution. Store critical insights as memory entries (type: 'code_fact' or 'pattern') using `local-memory-mcp` MCP tools `memory-store`.
31
31
  - **Commit**: Perform an atomic git commit and push for the changes made in the task.
32
- - **Handoff**: Use '@vheins/local-memory-mcp tools task-update' to document **detailed fix steps**, milestones, project-specific knowledge gained during execution, and validation evidence. If complex, decompose into smaller tasks using '@vheins/local-memory-mcp tools task-create'.
32
+ - **Handoff**: Use `local-memory-mcp` MCP tools `task-update` to document **detailed fix steps**, milestones, project-specific knowledge gained during execution, and validation evidence. If complex, decompose into smaller tasks using `local-memory-mcp` MCP tools `task-create`.
33
33
  - **Next**: Repeat this loop for the next 'pending' or 'stale' task.
34
- 5. **Backlog Migration**: Once all 'pending' and 'in_progress' tasks are completed or blocked, fetch tasks with status='backlog'. If any exist, select up to 20 tasks (prioritizing by priority field) and update their status to 'pending' using '@vheins/local-memory-mcp tools task-update' to ensure the next agent has work ready.
34
+ 5. **Backlog Migration**: Once all 'pending' and 'in_progress' tasks are completed or blocked, fetch tasks with status='backlog'. If any exist, select up to 20 tasks (prioritizing by priority field) and update their status to 'pending' using `local-memory-mcp` MCP tools `task-update` to ensure the next agent has work ready.
35
35
  6. **Report**: After processing all tasks, provide a summary of your progress.
36
36
 
37
37
  ## Mandatory Validation Rules
package/dist/mcp/prompts/definitions/tech-affinity-scout.md CHANGED
@@ -10,6 +10,6 @@ agent: Tech Scout
10
10
  I am working on the current repository using [{{tags}}].
11
11
 
12
12
  Please scout for relevant knowledge from other projects:
13
- 1. **Search**: Use '@vheins/local-memory-mcp tools memory-search' with current_tags=[{{tags}}] and include_archived=false.
13
+ 1. **Search**: Use `local-memory-mcp` MCP tools `memory-search` with current_tags=[{{tags}}] and include_archived=false.
14
14
  2. **Filter**: Look for 'patterns' or 'decisions' from other repositories that might apply here.
15
15
  3. **Translate**: Explain how these external best practices can be adapted to our current project context.
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@vheins/local-memory-mcp",
3
- "version": "0.5.13",
3
+ "version": "0.5.15",
4
4
  "description": "MCP Local Memory Service for coding copilot agents",
5
5
  "mcpName": "io.github.vheins/local-memory-mcp",
6
6
  "type": "module",