Allows capturing thoughts or tasks from an Android device via a floating bubble, widget, or quick capture button, which are then sent to the server for storage and retrieval.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@prospective-memory list open tasks"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Phone-first capture. MCP pull. Not Takeout.
Dump a thought in a few seconds on Android. Later ask Grok/Claude: “what’s open?”
No notifications in v1. No geofence. No Google history.
Sibling of google-activity-assistant — same kind of MCP host, different store.
Android (one box) --POST /v1/capture--> PC FastAPI + SQLite
│
Grok / Claude --MCP list_open_tasks----------┘Related MCP server: nudge
cd $env:USERPROFILE\Desktop\prospective-memory
uv sync
uv run pmem token # print X-PMEM-TOKEN for the phone
uv run pmem serve # http://0.0.0.0:8790Leave serve running. In another terminal / Grok, MCP serve-mcp is already wired if you add the config below.
Phone must reach this machine:
Same Wi‑Fi: http://YOUR_LAN_IP:8790
Out of house: Cloudflare Tunnel / Tailscale (see docs/V1.md)
Open android/ in Android Studio (or gradlew.bat assembleDebug).
Set server URL + token on first launch (same token as pmem token).
Save URL + token. Add lingo lines (d=dahi lena).
Enable floating + tile (overlay permission) — tap + → type → Enter. Drag the + onto the bottom ✕ to hide it; Enable turns it back on.
Home screen widget (stretch wide for a type-here bar) or long-press app icon → Capture. Keyboard opens immediately.
Empty Enter repeats the last lingo chip. Toast shows category · text. Offline posts queue and flush when the network is back.
Optional Write (handwriting test) is in android/.../ink/. First open downloads an on-device model. Not on the bubble. Drop it via ink/REMOVE.txt.
Allow HTTP cleartext for LAN IPs (already in network security config).
~/.grok/config.toml:
[mcp_servers.prospective_memory]
command = "C:\\Users\\bda99\\.local\\bin\\uv.exe"
args = [
"run",
"--directory",
"C:\\Users\\bda99\\Desktop\\prospective-memory",
"pmem",
"serve-mcp",
]
enabled = true
startup_timeout_sec = 60Ask: “list open tasks” / “anything grocery?” / “mark that dahi task done.”
uv run pmem capture "dahi lena"
uv run pmem list
uv run pmem statsOS reminders, geofence, WhatsApp check-in, glyphs, voice
Merging with Google Takeout activity
Multi-user cloud
Save a thought as an open task (same path as the Android capture box).
| Name | Required | Description | Default |
|---|---|---|---|
| text | Yes | ||
| source | No | mcp |
| Name | Required | Description |
|---|---|---|
No output parameters | ||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description must carry the behavioral burden. It states the core effect and adds the Android capture-box path, but does not disclose duplicate handling, failure modes, or whether any other state is changed.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
A single, front-loaded sentence with no filler. Every phrase contributes: action, result, and platform consistency.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
The tool is simple, has an output schema, and only one required parameter. Coverage of text plus the Android-capture context makes it callable; the main gap is the undocumented optional source.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 0%, so the description must compensate. 'Thought' gives a semantic hint for the required text param, but source is never mentioned, leaving it under-documented.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
States a specific action ('Save a thought as an open task') and a resource, and the Android capture-box reference adds useful context. This clearly differentiates it from sibling read/search/complete/drop operations.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The intended scenario is clear - save a thought as an open task - and the action itself distinguishes it from list/search/complete/drop. It does not name alternatives or exclusions, so it stops short of a 5.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Mark a task done.
| Name | Required | Description | Default |
|---|---|---|---|
| task_id | Yes |
| Name | Required | Description |
|---|---|---|
No output parameters | ||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden. It states the core status change but does not disclose whether the action is reversible, what happens when the task is already done, or how missing or invalid task IDs are handled.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is four words, front-loaded, and contains no filler. Every word adds meaning and directly states the tool's purpose.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
The tool is simple, has one parameter, and an output schema exists, so return value details are not strictly required. However, with no annotations and no usage or behavioral guidance, the description is only minimally complete for helping an agent select and invoke the tool correctly.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The only parameter, task_id, is self-explanatory from its name and title, but schema_description_coverage is 0% and the description adds no additional meaning. It does not say where the ID comes from, what format it uses, or whether it must reference an existing open task.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description uses a specific verb and object ('Mark a task done') that clearly states the tool's purpose and result. It differentiates this tool from siblings like capture_task, drop_task, and list_open_tasks by focusing on the completion status change.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
No guidance is given about when to use complete_task versus alternatives such as drop_task or capture_task. There is no mention of prerequisites, such as the task needing to already exist or be open, so an agent must infer usage from the tool name alone.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Drop a task without completing it.
| Name | Required | Description | Default |
|---|---|---|---|
| task_id | Yes |
| Name | Required | Description |
|---|---|---|
No output parameters | ||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden of behavioral disclosure. It states that a task is dropped, implying a mutating action, but it does not say whether this permanently deletes the task, whether it is reversible, or what state the task enters. For a destructive operation, this is a significant gap.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single concise sentence, front-loaded with the core verb. There is no redundant content. While it is short, it includes the essential differentiating phrase 'without completing it.'
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
An output schema exists, so return values are covered. However, for a mutating tool with no annotations, the description is too thin: it does not state the outcome for the task, whether it is removed from open lists, or any side effects. The tool's behavior and effects on task state are not adequately specified.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 0% and the description does not mention task_id. Although the parameter name and title are self-explanatory, the description adds no additional meaning beyond the schema. It fails to compensate for the missing parameter documentation.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description states a specific verb 'drop' and resource 'task', and the clause 'without completing it' differentiates it from complete_task. However, 'drop' is somewhat ambiguous—it could mean delete, cancel, or abandon—so it could be slightly more explicit about the exact operation.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies the appropriate use case: removing a task without marking it complete. It implicitly excludes complete_task, but it does not explicitly mention alternative sibling tools or spell out when not to use this tool. The context is clear enough for most agents.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
List open tasks, newest first. Optional category: grocery, bills, people, inbox, …
| Name | Required | Description | Default |
|---|---|---|---|
| limit | No | ||
| category | No |
| Name | Required | Description |
|---|---|---|
No output parameters | ||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
There are no annotations, so the description must carry behavioral disclosure. It adds useful behavior: tasks are open, sorting is newest first, and category is optional with open-ended examples. However, it does not state read-only guarantees, pagination/truncation behavior, or effects of the limit parameter.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is one efficient sentence with the core action and ordering front-loaded, followed by the optional category context. Every phrase adds value and there is no wasted wording.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given an output schema exists, return values do not need describing. The description covers the core list behavior and category filter, but it omits limit semantics and any guidance about when to choose this tool over search_tasks. For a simple tool this is adequate, but not fully complete.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 0%, so the description must compensate for parameter meaning. It adds examples for category ('grocery, bills, people, inbox, …') and says it is optional, but the limit parameter is entirely undocumentted in the description. With two parameters, only half gain real semantic explanation beyond the schema.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly specifies the verb 'List', the resource 'open tasks', and provides ordering ('newest first') plus an optional category filter with examples. It does not explicitly distinguish itself from the sibling tool search_tasks, so it meets the clarity bar without fully differentiating from alternatives.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
No guidance is given about when to use list_open_tasks versus search_tasks or other sibling tools. The category examples hint at filtering use, but there are no explicit when-to-use, when-not-to-use, or alternative-routing statements.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Search task text/category. By default only open items.
| Name | Required | Description | Default |
|---|---|---|---|
| limit | No | ||
| query | Yes | ||
| include_done | No |
| Name | Required | Description |
|---|---|---|
No output parameters | ||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the burden of behavioral disclosure. It does reveal an important default behavior ('By default only open items'), which directly informs the include_done parameter. But it does not explicitly state that search is read-only, describe result ordering/limits, or explain edge cases like empty queries. The read-only nature is arguably implied by the word 'Search'.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is two short sentences with no filler. The primary action is front-loaded, and the important default behavior is stated immediately after. Every word adds value.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
The output schema exists, so return value details are not needed. The description covers the core search semantics and the key default behavior. It could be slightly more explicit about the effect of include_done and limit, but those are reasonably inferable. For a low-complexity search tool, this is nearly complete.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The schema has no parameter descriptions (0% coverage), so the description must compensate. It clarifies that 'query' targets task text/category and implies what include_done does via the default statement. However, it does not explain the 'limit' parameter at all, leaving that to inference from its name and default value.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description states a specific verb ('Search'), a resource ('task'), and the fields being searched ('text/category'). This clearly distinguishes it from siblings like list_open_tasks (which presumably enumerates tasks) and task_stats (which reports statistics). The purpose is immediately understandable.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies usage: search tasks by text or category, with a default scope of open items. However, it does not explicitly mention alternatives or state when to prefer this over list_open_tasks or other siblings. The condition is only implied, not stated as guidance.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Counts by status and open categories.
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
| Name | Required | Description |
|---|---|---|
No output parameters | ||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden of behavioral disclosure. It reveals the aggregate nature (counts) but does not state whether counts include only open tasks, whether they cover a specific scope, or whether the operation has side effects. There is no annotation contradiction, but behavior is minimally described.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single short sentence with no filler, front-loading the core purpose ('Counts'). Every word earns its place.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a zero-parameter tool with an output schema, the description covers the essential grouping dimensions. The main gap is the ambiguity of 'open categories,' but the output schema likely resolves that once the tool is invoked.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The tool has zero parameters and schema description coverage is 100%, so there is no parameter semantics for the description to add. A baseline of 4 is appropriate because no parameter detail is needed.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description states a clear aggregation behavior: it counts tasks by status and by open categories, which is distinct from siblings that capture, list, search, complete, or drop tasks. However, 'open categories' is ambiguous and the resource is only implied by the tool name.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
No guidance is given on when to use this tool versus search_tasks or list_open_tasks, and no exclusions are mentioned. An agent cannot tell from the description whether to call this for a summary instead of a task list.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections. Dates show when Glama detected each change.
v0.5.0capture_taskcomplete_taskdrop_tasklist_open_taskssearch_taskstask_statsEach tool targets a distinct lifecycle action (capture, list, search, complete, drop, stats), so agents can generally tell them apart. The only mild overlap is that list_open_tasks and search_tasks can both retrieve open tasks by category, but search_tasks adds text matching to differentiate it.
Five of the six tools follow a clear verb_noun pattern: capture_task, list_open_tasks, search_tasks, complete_task, drop_task. task_stats breaks the pattern by using noun_noun, but this is a single minor deviation in an otherwise consistent set.
Six tools is well-scoped for a prospective-memory/task server. Each tool earns its place and there is no obvious redundancy or bloat.
The set covers the main task lifecycle: capture, list, search, complete, drop, and stats. The notable gap is the lack of an update/edit tool for changing task text or category, plus no explicit reopen flow, but these are workable gaps rather than fatal omissions.
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
Cloud-hosted MCP server for durable AI memory
Markdown-based note-taking with a hosted MCP server. Your notes serve you and your AI.
Google Keep-style notes app with an MCP server for AI agents to read/write notes.
Persistent personal memory for AI assistants — save, search, and recall across every MCP client.
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/Daemon-Killer/prospective-memory'
If you have feedback or need assistance with the MCP directory API, please join our Discord server
