# Strategic Agent Reasoning MCP (npm · @putervision/agent-reasoning-mcp)

Strategic BDI reasoning engine for autonomous AI agents — goals, utility, and replanning.

- Trust score: 75/100 (medium)
- Change this week: +4
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-09-25

## Components

- npm · `@putervision/agent-reasoning-mcp`: 75/100 (this document), [markdown](https://verifymcp.io/servers/putervision-agent-reasoning-mcp/putervision-agent-reasoning-mcp.md), [page](https://verifymcp.io/servers/putervision-agent-reasoning-mcp/putervision-agent-reasoning-mcp)

## Channel facts

- Registry: `npm`
- Package: `@putervision/agent-reasoning-mcp`
- Version: `0.2.1`
- Transport: `stdio`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically from public evidence about the published package, including repeated runs of it in an isolated sandbox, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-09-25.

- **Supply Chain Security**: 96/100
  - No malware found by supply-chain analysis.
  - No known CVEs affecting this package version or its production dependencies.
  - No install/post-install scripts declared.
  - 24 of 38 dependencies flagged as unhealthy (1 deprecated).
- **Provenance & Transparency**: 48/100
  - Source repository is publicly reachable at the declared URL.
  - Provenance check failed: no build-provenance attestation is published.
  - Clear OSI-approved license (MIT).
  - Actively maintained (last published 9 days ago).
  - Publishes a security disclosure policy (SECURITY.md).
- **Schema Quality & AI Usability**: 80/100
  - 100% of prompts and resources have a non-trivial description (not blank, and not just the item's name).
  - AI-judged instruction clarity (good).
  - Tool/resource definitions use about 1600 tokens (~76/item across 21 items; 10 tools + 11 resources), lean.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 50/100
  - Stability observed for 15 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 100/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 100% of tool parameters carry a description.
- **Tool Safety**: 100/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - We read all 10 captured tool definition(s), and no name or description among them implies an irreversible operation.
  - An AI judge read all 11 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 20/100
  - Spec-recency check failed: implements MCP spec 2024-11-05; the latest is 2026-07-28.

## Install

### How do I install the Strategic Agent Reasoning MCP server?

Strategic Agent Reasoning MCP runs locally as an npm package, launched with npx -y @putervision/agent-reasoning-mcp. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

### Claude

```bash
claude mcp add putervision-agent-reasoning-mcp -- npx -y @putervision/agent-reasoning-mcp
```

### Cursor

```json
{
  "mcpServers": {
    "putervision-agent-reasoning-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@putervision/agent-reasoning-mcp"
      ]
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "putervision-agent-reasoning-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@putervision/agent-reasoning-mcp"
      ]
    }
  }
}
```

### Codex

```bash
codex mcp add putervision-agent-reasoning-mcp -- npx -y @putervision/agent-reasoning-mcp
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "putervision-agent-reasoning-mcp": {
      "type": "local",
      "command": [
        "npx",
        "-y",
        "@putervision/agent-reasoning-mcp"
      ],
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add putervision-agent-reasoning-mcp --command npx --arg -y --arg @putervision/agent-reasoning-mcp
```

### Hermes

```yaml
mcp_servers:
  putervision-agent-reasoning-mcp:
    command: "npx"
    args: ["-y", "@putervision/agent-reasoning-mcp"]
```

### Netclaw

```json
{
  "McpServers": {
    "putervision-agent-reasoning-mcp": {
      "Transport": "stdio",
      "Command": "npx",
      "Arguments": [
        "-y",
        "@putervision/agent-reasoning-mcp"
      ]
    }
  }
}
```

### Vellum

```bash
assistant mcp add putervision-agent-reasoning-mcp -t stdio -c npx -a -y @putervision/agent-reasoning-mcp
```

### Other

```json
{
  "mcpServers": {
    "putervision-agent-reasoning-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@putervision/agent-reasoning-mcp"
      ]
    }
  }
}
```

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-09-25 (score 75, +1)

- [functional] We updated how we score, so this day's move reflects our rubric, not a change to the server

### 2026-09-23 (score 74, +1)

No change was recorded against any check on this day. Stability & Change Management went from 40 to 43. That category is still filling its 30-day observation window: 12 days of observed history at the previous scan, 13 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-09-21 (score 73, +1)

No change was recorded against any check on this day. Stability & Change Management went from 33 to 37. That category is still filling its 30-day observation window: 10 days of observed history at the previous scan, 11 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-09-19 (score 72, +1)

No change was recorded against any check on this day. Stability & Change Management went from 27 to 30. That category is still filling its 30-day observation window: 8 days of observed history at the previous scan, 9 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-09-17 (score 71, +16)

- [security improvement] Malware scan: unverified → pass

### 2026-09-16 (score 55, −12)

- [security regression] Malware scan: pass → unverified
- [functional regression] Dependency health: 0.81 → 0.68
- [functional regression] Schema quality: 900 → 1600
- [functional regression] MCP protocol: pass → fail
- [functional improvement] Tool coverage: 0% → 100%
- [functional improvement] Stability: unverified → 0.20
- [functional] Schema quality: fair → good
- [functional] Package version: 0.1.2 → 0.2.1

### 2026-09-12 (score 67, +15)

- [security improvement] Malware scan: unverified → pass

### 2026-09-10 (score 52)

First indexed and scored.

## MCP tools (10)

### `set_goal` (~241 tokens)

Register, update, decompose, or manage hierarchical goals and task DAGs in the reasoning engine.

Input parameters:

- `action` (string, required): The goal management operation to perform
- `client_request_id` (string): Idempotency key to prevent duplicate creation
- `deadline_at` (string): ISO-8601 deadline timestamp
- `description` (string): Detailed goal description
- `id` (string): Goal ID (required for update, get, abandon)
- `limit` (number): Max items to return for list action
- `parent_id` (string): Parent goal ID for hierarchical sub-goals
- `priority` (number): Goal priority (0.0 to 1.0)
- `progress` (number): Completion progress (0.0 to 1.0)
- `project` (string): Target project slug
- `status` (string): Goal status
- `subgoals` (array): Array of sub-goals for decompose action
- `success_criteria` (array): List of verifiable conditions
- `title` (string): Goal title or objective summary
- `utility_weights` (object): Goal-specific utility weight overrides

### `evaluate_situation` (~179 tokens)

Ingest multi-modal situation snapshot, compute expected utilities against active weights, and output prioritized action recommendations.

Input parameters:

- `action` (string, required): Snapshot evaluation mode or quick text context
- `candidate_actions` (array): Candidate actions to score and rank
- `lookahead_depth` (number): Bounded heuristic lookahead plies (e.g. 2-3 plies, discount gamma=0.85)
- `project` (string): Target project slug
- `quick_context` (string): Text summary of current situation for quick evaluation
- `session_id` (string): Linked state-memory session ID
- `snapshot` (object): Normalized SituationSnapshot with world, vision, state, and vitals
- `trace_id` (string): Distributed trace ID
- `utility_profile` (string): Named utility profile to score against (defaults to active)

### `replan` (~105 tokens)

Regenerate sub-task DAG and abort/recreate intentions upon unexpected blockers or environmental state changes.

Input parameters:

- `action` (string, required): Replanning trigger type
- `blocker_description` (string): Description of the obstacle or blocker encountered
- `goal_id` (string, required): ID of goal to replan
- `preserve_completed` (boolean): Whether to preserve already completed subgoals
- `project` (string): Target project slug
- `trigger_event` (string): Event description triggering replanning

### `assess_risk` (~94 tokens)

Compute quantitative risk and threat assessment for candidate actions or plans against active utility weights.

Input parameters:

- `action` (string, required): Risk assessment mode
- `candidate_action` (string): Action name to evaluate
- `candidate_actions` (array): Multiple actions to compare risk scores
- `parameters` (object): Action parameters
- `project` (string): Target project slug
- `situation_context` (object): Current environment telemetry & vitals

### `query_knowledge` (~90 tokens)

Search learned heuristic patterns, tactics, and past decision traces by context similarity.

Input parameters:

- `action` (string, required): Knowledge query mode
- `context_tags` (array): Filter by context tags
- `limit` (number): Max patterns to return
- `pattern_type` (string): Filter by pattern category
- `project` (string): Target project slug
- `query` (string): Semantic search query string

### `set_utility_weights` (~120 tokens)

Configure and activate multi-attribute utility weights (aggression, caution, greed, exploration, cooperation).

Input parameters:

- `action` (string, required): Profile operation
- `description` (string): Profile description
- `is_active` (boolean): Whether to set as currently active profile
- `name` (string): Profile name (e.g. "aggressive", "cautious", "explorer")
- `project` (string): Target project slug
- `weights` (object): Key-value map of weight values (0.0 to 1.0)

### `get_decision_trace` (~92 tokens)

Retrieve explainable step-by-step chain-of-thought rationale, candidate utilities, and risk assessment for past decisions.

Input parameters:

- `action` (string, required): Trace retrieval operation
- `goal_id` (string): Filter traces by linked goal ID
- `limit` (number): Max traces to list
- `project` (string): Target project slug
- `trace_id` (string): Trace ID for get/explain action

### `manage_beliefs` (~195 tokens)

Maintain structured belief state with TTL expiration sweeps, exponential confidence decay, and category filtering.

Input parameters:

- `action` (string, required): Belief operation
- `belief_id` (string): Belief ID for specific lookup
- `category` (string): Belief category
- `client_request_id` (string): Idempotency key
- `confidence` (number): Confidence score (0.0 to 1.0)
- `decay_rate` (number): Exponential decay rate lambda per hour
- `expires_at` (string): ISO-8601 expiration timestamp
- `object`: Belief value / state payload
- `predicate` (string): Predicate relationship (e.g. "is_locked", "status")
- `project` (string): Target project slug
- `source` (string): Belief provenance source
- `subject` (string): Belief subject (e.g. "north_gate", "enemy_patrol")

### `manage_intentions` (~205 tokens)

Queue, dispatch, track, and resolve behavior directives (wire contract) for behavior-mcp.

Input parameters:

- `abort_conditions` (array): Auto-abort trigger conditions
- `action` (string, required): Intention operation
- `behavior_name` (string): Target behavior tree name (e.g. "combat_kite", "gather_loop")
- `client_request_id` (string): Idempotency key
- `deadline_at` (string): ISO-8601 completion deadline
- `goal_id` (string): Linked goal ID
- `intention_id` (string): Intention ID for dispatch/get/cancel/resolve
- `parameters` (object): Runtime behavior parameters
- `priority` (number): Execution priority (0.0 to 1.0)
- `project` (string): Target project slug
- `result` (object): Outcome payload for resolve action
- `status` (string): Status filter or update
- `trace_id` (string): Linked decision trace ID

### `manage_reasoning_db` (~86 tokens)

Database maintenance, stats, SHA-256 Merkle audit verification, checkpoints save/restore, and diffs.

Input parameters:

- `action` (string, required): Database maintenance operation: stats, audit, doctor (health diagnostics), snapshot, diff, restore
- `description` (string): Description for snapshot
- `name` (string): Snapshot name
- `project` (string): Target project slug

## Diagnostics

Captured diagnostic sections: Provenance, Dependencies. The full working is on the page: https://verifymcp.io/servers/putervision-agent-reasoning-mcp/putervision-agent-reasoning-mcp#diagnostics

## Score history

- 2026-09-25: 75
- 2026-09-24: 74
- 2026-09-23: 74
- 2026-09-22: 73
- 2026-09-21: 73
- 2026-09-20: 72
- 2026-09-19: 72
- 2026-09-18: 71
- 2026-09-17: 71
- 2026-09-16: 55
- 2026-09-15: 67
- 2026-09-14: 67
- 2026-09-13: 67
- 2026-09-12: 67
- 2026-09-11: 52
- 2026-09-10: 52

## Common questions

### What is the Strategic Agent Reasoning MCP server?

Strategic Agent Reasoning MCP is listed in the public MCP registry as io.github.putervision/agent-reasoning-mcp. Strategic BDI reasoning engine for autonomous AI agents, goals, utility, and replanning. This page covers its npm package (@putervision/agent-reasoning-mcp).

### Is the Strategic Agent Reasoning MCP server safe to use?

Strategic Agent Reasoning MCP scores 75 out of 100 on VerifyMCP. We found no known CVEs affecting it as of 25 September 2026. It declares no install or post-install scripts. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

### What tools does the Strategic Agent Reasoning MCP server expose?

Strategic Agent Reasoning MCP exposes 10 tools: set_goal, evaluate_situation, replan, assess_risk, query_knowledge, and 5 more. Their descriptions and schemas cost roughly 1,407 tokens of context every time the server is loaded.

### Is the Strategic Agent Reasoning MCP server still maintained?

Strategic Agent Reasoning MCP is still listed as active in the MCP registry. We last reached this channel on 25 September 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

### What licence is the Strategic Agent Reasoning MCP server under?

Strategic Agent Reasoning MCP declares the MIT licence, which is OSI-approved. That covers the source only, and says nothing about the cost of any service it calls.

## Links

- npm package: https://www.npmjs.com/package/@putervision/agent-reasoning-mcp
- Socket report: https://socket.dev/npm/package/@putervision/agent-reasoning-mcp
- Repository: https://github.com/putervision/agent-reasoning-mcp
- Website: https://agentreasoningmcp.com/
- Changelog RSS feed: https://verifymcp.io/servers/putervision-agent-reasoning-mcp/putervision-agent-reasoning-mcp.xml
- Changelog JSON feed: https://verifymcp.io/servers/putervision-agent-reasoning-mcp/putervision-agent-reasoning-mcp.json
- HTML version of this page: https://verifymcp.io/servers/putervision-agent-reasoning-mcp/putervision-agent-reasoning-mcp