// SCAN REPORT

Claude Flow

454 tools. 242 can modify or destroy data without limits.

28 destructive tools with no built-in limits. Policy required.

Last updated:

242 can modify or destroy data
212 read-only
454 tools total

Community server · catalogue entry verified 26/06/2026

How to control Claude Flow ↓

TOOL MAP

What Claude Flow exposes to your agents

Read (212) Write / Execute (212) Destructive / Financial (28)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS
Critical Risk

The most dangerous Claude Flow tools

Financial · Medium federation_report_spend This tool explicitly reports financial spend/cost data to a cost-tracking system. Financial · Critical hooks_transfer Transfer learned patterns from another project Use when native Bash hooks (via Claude Code\ Financial · Critical transfer_ipfs-resolve Resolve IPNS name to CID Use when native package install ( Destructive · Critical system_reset A system reset that wipes or reinitializes live daemon state (neural indices, reasoning banks, swarm configurations, circuit breaker states) is effectively irreversible without a prior snapshot or backup.

242 of Claude Flow's 454 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control Claude Flow

PolicyLayer is an MCP gateway — it sits between your AI agents and Claude Flow, and nothing reaches the server without passing your rules. These are the rules we recommend:

Block financial tools by default
{
  "hooks_transfer": {
    "deny_if": [
      {
        "conditions": [],
        "on_deny": "Requires human approval."
      }
    ]
  }
}

Financial tools should be explicitly enabled per use case, not open by default.

Deny destructive operations
{
  "agentdb_batch": {
    "deny_if": [
      {
        "conditions": [],
        "on_deny": "Blocked by default. Requires approval."
      }
    ]
  }
}

Destructive tools should never be available to autonomous agents without human approval.

Rate limit write operations
{
  "gt_beads_dep": {
    "limits": [
      {
        "counter": "gt_beads_dep_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations
{
  "claims_board": {
    "limits": [
      {
        "counter": "claims_board_per_minute",
        "window": "minute",
        "max": 60,
        "scope": "grant"
      }
    ]
  }
}

Controls API costs and prevents retry loops from exhausting upstream rate limits.

  1. Create a free account and register Claude Flow — nothing to install.
  2. Add these rules — paste them, or build them visually. Tune the limits to your setup.
  3. Point your MCP client (Claude, Cursor, anything) at your gateway URL.
ENFORCE POLICY ON CLAUDE FLOW →

Instant setup, no code required.

FULL CATALOGUE

All 454 Claude Flow tools

FINANCIAL 3 tools
Financial hooks_transfer Transfer learned patterns from another project Use when native Bash hooks (via Claude Code\ Financial transfer_ipfs-resolve Resolve IPNS name to CID Use when native package install ( Financial federation_report_spend ADR-097 Phase 3 upstream: report the actual cost of a completed federated call. Fans out to the cost-tracker b
DESTRUCTIVE 25 tools
Destructive agentdb_batch Batch operations on AgentDB episodes (insert, update, delete). Note: entries are stored in the AgentDB episode Destructive agentdb_causal-edge-delete Delete a causal edge between two memory entries. Returns controller= Destructive agentdb_causal-node-delete Cascade-delete a causal node and all its incident edges from the SQL fallback. Native graph-node entries are u Destructive agentdb_hierarchical-delete Delete a hierarchical-memory entry by key. Returns controller= Destructive autopilot_reset Reset autopilot iteration counter and restart the timer. Use when running long-horizon goals that should resum Destructive config_reset Reset configuration to defaults Use when native settings.json edits are wrong because the values need to be re Destructive hooks_intelligence-reset Reset intelligence learning state Use when native Bash hooks (via Claude Code\ Destructive hooks_worker-cancel Cancel a running worker Use when native Bash hooks (via Claude Code\ Destructive iot_device_remove Remove a registered device Destructive iot_fleet_delete Delete a fleet Destructive iot_fleet_remove_device Remove a device from a fleet Destructive memory_delete Remove a stored memory entry by exact (namespace, key). Use when a previously stored decision is invalidated o Destructive ruvector_delete Delete vectors by ID. Destructive session_delete Delete a saved session Use when native conversation memory is wrong because you need durable cross-session sta Destructive system_reset Reset system state Use when native Bash is wrong because you need Ruflo runtime metrics (HNSW index size, Reas Destructive task_cancel Cancel a task Use when native TodoWrite is wrong because you need cross-session task persistence, agent assign Destructive wasm_agent_reset Reset a WASM agent — clears messages and turn count so it can be reused across tasks. Use when native Task is Destructive wasm_gallery_remove_custom Remove a custom template from the WASM gallery by ID. Use when native Bash rm is wrong because custom template Destructive workflow_cancel Cancel a workflow Use when native TodoWrite + sequential Bash is wrong because the work has a real dependency Destructive workflow_delete Delete a workflow Use when native TodoWrite + sequential Bash is wrong because the work has a real dependency Destructive iot_firmware_rollout_rollback Force rollback a firmware rollout Destructive agent_terminate Remove a Ruflo-tracked agent from the registry and free its swarm slot. Use when you need to (a) clean up a sp Destructive federation_wg_keyrotate ADR-111 Phase 6: rotate the local WG keypair. Returns the new public key + recommended next steps (republish m Destructive managed_agent_terminate Delete a managed cloud-agent session (stops billing for it) — the CLOUD counterpart of wasm_agent_terminate. U Destructive teammate_cleanup Cleanup team resources and save state
EXECUTE 121 tools
Execute agent_pool Manage a fixed-size warm pool of pre-spawned agents to skip cold-start cost on bursty workloads. Use when nati Execute agentdb_consolidate Run memory consolidation to promote entries across tiers and compress old data Use when generic memory_ tools Execute benchmark_run Execute performance benchmarks (V2 compatible). Deprecated: Use system/metrics instead. Execute browser_eval Execute JavaScript in page context Use when native WebFetch is wrong because you need real browser automation Execute browser_wait Wait for a condition Use when native WebFetch is wrong because you need real browser automation — JS-heavy SPA Execute claims_accept-handoff Accept a pending handoff Use when nothing native covers per-agent capability gating — Claude Code agents have Execute claims_claim Claim an issue for work (human or agent) Use when nothing native covers per-agent capability gating — Claude C Execute claims_handoff Request handoff of an issue to another claimant Use when nothing native covers per-agent capability gating — C Execute claims_rebalance Suggest or apply load rebalancing across agents Use when nothing native covers per-agent capability gating — C Execute claims_release Release a claim on an issue Use when nothing native covers per-agent capability gating — Claude Code agents ha Execute claims_steal Steal a stealable issue Use when nothing native covers per-agent capability gating — Claude Code agents have f Execute daa_agent_adapt Trigger agent adaptation based on feedback Use when native Task is wrong because you need agents that adapt th Execute daa_workflow_execute Execute a DAA workflow Use when native Task is wrong because you need agents that adapt their cognitive patter Execute embeddings_rabitq_build Build RaBitQ 1-bit quantized index from stored embeddings (32× compression). Pre-filters candidates via Hammin Execute execute-workflow Execute a workflow Execute gt_formula_execute Execute a formula to create beads/molecules in Gas Town Execute gt_wasm_parse_formula Parse TOML formula content to AST using WASM (352x faster than JavaScript) Execute hooks_build-agents Generate optimized agent configurations from pretrain data Use when native Bash hooks (via Claude Code\ Execute hooks_intelligence_trajectory-end End trajectory and trigger SONA learning with EWC++ Use when native Bash hooks (via Claude Code\ Execute hooks_intelligence_trajectory-start Begin SONA trajectory for reinforcement learning Use when native Bash hooks (via Claude Code\ Execute hooks_pre-task Record task start and get agent suggestions with intelligent model routing (ADR-026) Use when native Bash hook Execute hooks_session-start Initialize a new session and auto-start daemon Use when native Bash hooks (via Claude Code\ Execute hyperbolic_entailment_graph Build and query entailment graphs using hyperbolic embeddings. Supports transitive closure and pruning strateg Execute performance_benchmark Run performance benchmarks Use when native shell timing ( Execute performance_optimize Apply performance optimizations Use when native shell timing ( Execute performance_profile Profile specific component or operation Use when native shell timing ( Execute ruvector-attention Execute attention mechanism (39 types: multi-head, flash, sparse, linear, etc.) Execute ruvector-gnn Execute GNN layer (GCN, GAT, GraphSAGE, GIN, MPNN, EdgeConv) Execute ruvllm_sona_adapt Run SONA instant adaptation with a quality signal. Use when sending every prompt to the Anthropic API is wrong Execute rvf_build Build RVF container Execute teammate_launch_swarm Launch swarm to execute an approved plan Execute terminal_execute Execute a command in a terminal session Use when native Bash is wrong because you need a persistent terminal s Execute wasm_agent_prompt Send a prompt to a WASM agent and get a response. Use when native Task is wrong because the workload needs san Execute wasm_agent_tool Execute a tool on a WASM agent sandbox. Tools: read_file, write_file, edit_file, write_todos, list_files. Use Execute workflow_execute Execute a workflow Use when native TodoWrite + sequential Bash is wrong because the work has a real dependency Execute workflow_run Run a workflow from a template or file Use when native TodoWrite + sequential Bash is wrong because the work h Execute agent_spawn Spawn a subagent Execute autopilot_predict Predict the optimal next action based on current state and learned patterns. Use when running long-horizon goa Execute browser_back Navigate back in browser history Use when native WebFetch is wrong because you need real browser automation — Execute browser_check Check a checkbox Use when native WebFetch is wrong because you need real browser automation — JS-heavy SPA scr Execute browser_forward Navigate forward in browser history Use when native WebFetch is wrong because you need real browser automation Execute browser_hover Hover over an element using ref (@e1) or CSS selector Use when native WebFetch is wrong because you need real Execute browser_reload Reload the current page Use when native WebFetch is wrong because you need real browser automation — JS-heavy Execute browser_session_end End a recorded browser session: trajectory-end with verdict, rvf compact, AIDefence pre-store gate (best-effor Execute browser_session_record Open a named, traced browser session: allocate an RVF cognitive container, begin a ruvector trajectory, then o Execute browser_uncheck Uncheck a checkbox Use when native WebFetch is wrong because you need real browser automation — JS-heavy SPA s Execute collective-decide Request collective decision from agents Execute coordination_orchestrate Orchestrate multi-agent coordination Use when native Task is wrong because the work crosses multiple agents th Execute coordination_sync Synchronize state across nodes Use when native Task is wrong because the work crosses multiple agents that nee Execute daa_cognitive_pattern Analyze or change cognitive patterns Use when native Task is wrong because you need agents that adapt their co Execute embeddings_init Initialize the ONNX embedding subsystem with hyperbolic support Use when text similarity matters beyond keywor Execute emit-event Emit an event to subscribers Execute federation_consensus Propose a federated consensus operation across all active peers Execute federation_init Initialize federation on this node with a manifest and begin discovery Execute gt_formula_cook Cook a formula into a protomolecule with variable substitution (WASM accelerated) Execute gt_wasm_cook_batch Batch cook multiple formulas using WASM for maximum performance Execute gt_wasm_optimize_convoy Optimize convoy execution order using WASM graph algorithms Execute hooks_coverage-route Route task to agents based on test coverage gaps (ruvector integration) Execute hooks_intelligence_attention Compute attention-weighted similarity using MoE/Flash/Hyperbolic Use when native Bash hooks (via Claude Code\ Execute hooks_intelligence_learn Force immediate SONA learning cycle with EWC++ consolidation Use when native Bash hooks (via Claude Code\ Execute hooks_model-route Route task to optimal Claude model (haiku/sonnet/opus) based on complexity Use when native Bash hooks (via Cla Execute hooks_pre-command Assess risk before executing a command Use when native Bash hooks (via Claude Code\ Execute hooks_pretrain Analyze repository to bootstrap intelligence (4-step pipeline) Use when native Bash hooks (via Claude Code\ Execute hooks_teammate-idle Agent Teams hook — fired when a teammate agent finishes its turn; reports whether a pending task can be auto-a Execute hooks_worker-dispatch Dispatch a background worker for analysis/optimization tasks Use when native Bash hooks (via Claude Code\ Execute hyperbolic_embed_hierarchy Embed hierarchical structure in Poincare ball. Uses hyperbolic geometry for optimal tree representation with l Execute iot_firmware_rollout_advance Advance a firmware rollout to next stage Execute neural_optimize Optimize neural model performance Use when nothing native trains on your workflow — Claude Code has no learnin Execute neural_predict Make predictions using a neural model Use when nothing native trains on your workflow — Claude Code has no lea Execute neural_train Train a neural model Use when nothing native trains on your workflow — Claude Code has no learning loop. Use t Execute pr_causal_infer Perform causal inference using do-calculus Execute quantum_annealing_solve Solve combinatorial optimization using quantum annealing simulation. Supports QUBO, Ising, SAT, Max-Cut, TSP, Execute quantum_grover_search Grover-inspired search with quadratic speedup for unstructured search problems. Provides O(sqrt(N)) query comp Execute quantum_qaoa_optimize Optimize using Quantum Approximate Optimization Algorithm. Best for Max-Cut, portfolio optimization, schedulin Execute ruvector-hyperbolic Hyperbolic embedding operations (Poincare ball, Lorentz hyperboloid) Execute ruvector-optimize Self-learning query optimization and index tuning Execute ruvllm_chat_format Format chat messages using a template (llama3, mistral, chatml, phi, gemma, or auto-detect). Use when sending Execute ruvllm_microlora_adapt Adapt MicroLoRA weights with quality feedback. Use when sending every prompt to the Anthropic API is wrong bec Execute swarm_init Initialize a swarm with persistent state tracking Use when native Task tool is wrong because you need multi-ag Execute task_orchestrate Orchestrate a task Execute teammate_batch_route Route multiple tasks to teammates optimally, avoiding over-assignment. Execute teammate_route_task Route a task to the best-suited teammate using semantic matching Execute teammate_spawn Spawn a new teammate in a team. Returns AgentInput for Claude Code Task tool. Execute teammate_teleport Teleport team to a new context/working directory Execute workflow_pause Pause a running workflow Use when native TodoWrite + sequential Bash is wrong because the work has a real depe Execute agentdb_route Route a task via AgentDB SemanticRouter or LearningSystem recommendAlgorithm Use when generic memory_ tools ar Execute agentdb_semantic-route Route an input via AgentDB SemanticRouter for intent classification Use when generic memory_ tools are wrong b Execute autopilot_disable Disable autopilot. Agents will be allowed to stop even if tasks remain. Use when running long-horizon goals th Execute autopilot_enable Enable autopilot persistent completion. Agents will be re-engaged when tasks remain incomplete. Use when runni Execute browser_click Click an element using ref (@e1) or CSS selector Use when native WebFetch is wrong because you need real brows Execute browser_close Close the browser session Use when native WebFetch is wrong because you need real browser automation — JS-heav Execute browser_fill Clear and fill an input element Use when native WebFetch is wrong because you need real browser automation — J Execute browser_open Navigate browser to a URL Use when native WebFetch is wrong because you need real browser automation — JS-heav Execute browser_press Press a keyboard key Use when native WebFetch is wrong because you need real browser automation — JS-heavy SPA Execute browser_scroll Scroll the page Use when native WebFetch is wrong because you need real browser automation — JS-heavy SPA scra Execute browser_select Select an option from a dropdown Use when native WebFetch is wrong because you need real browser automation — Execute browser_template_apply Fetch a recipe from the browser-templates AgentDB namespace and return it for caller-level execution. Use when Execute browser_type Type text with key events (for autocomplete, etc.) Use when native WebFetch is wrong because you need real bro Execute coordination_consensus Manage consensus protocol with BFT, Raft, or Quorum strategies Use when native Task is wrong because the work Execute coordination_node Manage coordination nodes Use when native Task is wrong because the work crosses multiple agents that need to Execute daa_agent_create Create a decentralized autonomous agent Use when native Task is wrong because you need agents that adapt their Execute federation_send Send a message to a federated peer through the PII pipeline and security gates. Optional budget controls (ADR- Execute github_workflow Manage GitHub Actions workflows Use when native Bash / file tools are wrong because this MCP tool exposes Rufl Execute gt_wasm_resolve_deps Resolve bead dependencies using WASM (topological sort, cycle detection, critical path) Execute hive-mind_broadcast Broadcast message to all workers Use when native Task is wrong because you need queen-led collective intellige Execute hive-mind_consensus Propose or vote on consensus with BFT, Raft, or Quorum strategies Use when native Task is wrong because you ne Execute hive-mind_init Initialize the hive-mind collective Use when native Task is wrong because you need queen-led collective intell Execute hive-mind_join Join an agent to the hive-mind Use when native Task is wrong because you need queen-led collective intelligenc Execute hive-mind_spawn Spawn workers and automatically join them to the hive-mind (combines agent/spawn + hive-mind/join) Use when na Execute managed_agent_create Spin up an Anthropic-managed cloud agent (Agent + Environment + Session) — the CLOUD counterpart of wasm_agent Execute managed_agent_prompt Send a user turn to a managed cloud-agent session and wait for it to go idle, returning the assistant text + a Execute memory_migrate Manually trigger migration from legacy JSON store to sql.js Use when native Read/Write is wrong because you ne Execute quantum_dependency_resolve Resolve complex dependency graphs using quantum-inspired optimization. Handles version conflicts, minimizes pa Execute quantum_schedule_optimize Optimize task scheduling using quantum algorithms. Minimizes makespan, cost, or maximizes resource utilization Execute ruvllm_sona_create Create a SONA instant adaptation loop (<1ms adaptation cycles). Use when sending every prompt to the Anthropic Execute teammate_enable_optimizers Enable BMSSP-powered optimization for team topology and task routing. Execute teammate_push_remote Push team to Claude.ai remote session Execute terminal_create Create a new terminal session Use when native Bash is wrong because you need a persistent terminal session acr Execute wasm_agent_create Create a sandboxed WASM agent with virtual filesystem (no OS access). Optionally use a gallery template. Use w Execute wasm_gallery_create Create a WASM agent from a gallery template. Use when native Task is wrong because the workload needs sandboxe Execute workflow_resume Resume a paused workflow Use when native TodoWrite + sequential Bash is wrong because the work has a real depe
WRITE 91 tools
Write gt_beads_dep Add or remove dependencies between beads Write gt_convoy_track Add or remove issues from an existing convoy Write hive-mind_leave Remove an agent from the hive-mind Use when native Task is wrong because you need queen-led collective intelli Write agentdb_session-end End session, persist to ReflexionMemory, trigger NightlyLearner consolidation Use when generic memory_ tools a Write agentdb_session-start Start a session with ReflexionMemory episodic replay Use when generic memory_ tools are wrong because you need Write aidefence_learn Record detection feedback for pattern learning. Improves future detection accuracy. Use when nothing native ex Write claims_status Update claim status Use when nothing native covers per-agent capability gating — Claude Code agents have file- Write daa_knowledge_share Share knowledge between agents Use when native Task is wrong because you need agents that adapt their cognitiv Write federation_wg_attest ADR-111 Phase 6 (witness chain): emit an operator-signed attestation entry for a coordination change. Returns Write feedback_record Record feedback for learning Write gt_beads_sync Synchronize beads between Gas Town and Claude Flow AgentDB Write hooks_init Initialize hooks in project with .claude/settings.json Use when native Bash hooks (via Claude Code\ Write hooks_intelligence_pattern-store Store pattern in ReasoningBank (HNSW-indexed) Use when native Bash hooks (via Claude Code\ Write hooks_model-outcome Record model routing outcome for learning Use when native Bash hooks (via Claude Code\ Write hooks_task-completed Agent Teams hook — fired when a task is marked complete; records completion and (eventually) trains patterns + Write iot_store_ingest Ingest vectors into device store Write memory_store Store information in semantic memory Write pattern_store Store learned pattern Write pr_memory_gate Pre-storage coherence gate for memory entries Write progress_sync Calculate and persist V3 progress metrics to file Use when native TodoWrite is wrong because you need cross-se Write store-reasoning Store a reasoning pattern Write teammate_broadcast Broadcast message to all teammates in a team Write teammate_share_transcript Share message transcript with another teammate Write witness_append Append to witness chain Write witness_log Log to witness chain (immutable audit) Write agent_update Mutate a tracked agent\ Write agentdb_causal-edge Record a causal edge between two memory entries via CausalMemoryGraph Use when generic memory_ tools are wrong Write agentdb_feedback Record task feedback for learning via LearningSystem + ReasoningBank controllers Use when generic memory_ tool Write agentdb_hierarchical-store Store to hierarchical memory with tier (working, episodic, semantic) Use when generic memory_ tools are wrong Write agentdb_pattern-store Store a pattern directly via ReasoningBank controller Use when generic memory_ tools are wrong because you nee Write autopilot_config Configure autopilot limits: max iterations (1-1000), timeout in minutes (1-1440), and task sources. Use when r Write claims_mark-stealable Mark an issue as stealable by other agents Use when nothing native covers per-agent capability gating — Claude Write config_import Import configuration from JSON Use when native settings.json edits are wrong because the values need to be rea Write config_set Set configuration value Use when native settings.json edits are wrong because the values need to be read by th Write coordination_topology Configure swarm topology Use when native Task is wrong because the work crosses multiple agents that need to v Write create-workflow Create a new workflow Write daa_workflow_create Create an autonomous workflow Use when native Task is wrong because you need agents that adapt their cognitive Write git-commit Create a Git commit Write github_issue_track Track and manage issues Use when native Bash / file tools are wrong because this MCP tool exposes Ruflo-specif Write github_pr_manage Manage pull requests Use when native Bash / file tools are wrong because this MCP tool exposes Ruflo-specific Write gt_beads_create Create a bead (issue/task) in the Gas Town Beads system with priority and labels Write gt_convoy_create Create a convoy (work order) for tracking and coordinating multiple beads Write gt_formula_create Create a custom formula definition Write gt_mail Send, read, or list Gas Town internal mail messages Write gt_sling Sling (assign) a bead to a Gas Town agent for processing Write hive-mind_shutdown Shutdown the hive-mind and terminate all workers Use when native Task is wrong because you need queen-led coll Write hooks_notify Send cross-agent notification Use when native Bash hooks (via Claude Code\ Write hooks_post-command Record command execution outcome Use when native Bash hooks (via Claude Code\ Write hooks_post-edit Record editing outcome for learning Use when native Bash hooks (via Claude Code\ Write hooks_post-task Record task completion for learning Use when native Bash hooks (via Claude Code\ Write hooks_session-restore Restore a previous session Use when native Bash hooks (via Claude Code\ Write iot_device_register Register a Cognitum Seed device by endpoint Write iot_firmware_rollout_create Create a firmware rollout for a fleet Write iot_fleet_add_device Add a registered device to a fleet Write iot_fleet_create Create a new device fleet Write memory_import_claude Import Claude Code auto-memory files into AgentDB with ONNX vector embeddings. Reads ~/.claude/projects//memor Write memory_usage Manage coordination memory (V2 compatible). Deprecated: Use memory/store, memory/search, or memory/list instea Write neural_patterns Get or manage neural patterns Use when nothing native trains on your workflow — Claude Code has no learning lo Write post-browse Record browser automation outcome for learning Write results_merge Merge results with CRDT Write ruvector_create_index Create a vector index (HNSW or IVF) for faster similarity search. Write ruvector_insert Insert vectors into a table. Supports batch insertion and upsert. Write ruvector_update Update an existing vector and/or its metadata. Write ruvllm_generate_config Create a generation config (maxTokens, temperature, topP, etc.) as JSON. Use when sending every prompt to the Write ruvllm_hnsw_add Add a pattern to an HNSW router. Embedding must match router dimensions. Use when sending every prompt to the Write ruvllm_hnsw_create Create a WASM HNSW router for semantic pattern routing. Max ~11 patterns (v2.0.1 limit). Use when sending ever Write ruvllm_microlora_create Create a MicroLoRA adapter (ultra-lightweight LoRA, ranks 1-4). Use when sending every prompt to the Anthropic Write session_restore Restore a saved session Use when native conversation memory is wrong because you need durable cross-session st Write session_save Save current session state Use when native conversation memory is wrong because you need durable cross-session Write swarm_shutdown Shutdown a swarm and update persistent state Use when native Task tool is wrong because you need multi-agent c Write task_assign Assign a task to one or more agents Use when native TodoWrite is wrong because you need cross-session task per Write task_complete Mark task as complete Use when native TodoWrite is wrong because you need cross-session task persistence, agen Write task_create Create a new task Use when native TodoWrite is wrong because you need cross-session task persistence, agent as Write task_update Update task status or progress Use when native TodoWrite is wrong because you need cross-session task persiste Write teammate_approve_plan Approve a submitted plan Write teammate_save_memory Save teammate session memory to disk Write teammate_send_message Send a message to a specific teammate Write teammate_spawn_team Create a new team for multi-agent collaboration using native TeammateTool. Write teammate_submit_plan Submit a plan for team approval Write teammate_update_context Update team shared context Write terminal_close Close a terminal session Use when native Bash is wrong because you need a persistent terminal session across t Write todo_add Add a task to the todo list Write todo_complete Mark a task as complete Write wasm_agent_export Export a WASM agent\ Write wasm_agent_terminate Terminate a WASM agent and free resources. Use when native Task is wrong because the workload needs sandboxed Write wasm_gallery_add_custom Add a custom agent template to the WASM gallery registry. Use when native Write is wrong because custom templa Write wasm_gallery_configure Apply runtime configuration overrides (e.g. maxTurns, model) to the active WASM gallery template. Use when nat Write witness_export Export chain for external audit Write workflow_create Create a new workflow Use when native TodoWrite + sequential Bash is wrong because the work has a real depende Write workflow_template Save workflow as template or create from template Use when native TodoWrite + sequential Bash is wrong because Write write_file Write a file to the virtual filesystem (worker mock)
READ 212 tools
Read claims_board Get a visual board view of all claims Use when nothing native covers per-agent capability gating — Claude Code Read claims_stealable List all stealable issues Use when nothing native covers per-agent capability gating — Claude Code agents have Read daa_performance_metrics Get DAA performance metrics Use when native Task is wrong because you need agents that adapt their cognitive p Read performance_bottleneck Detect performance bottlenecks Use when native shell timing ( Read performance_metrics Get detailed performance metrics Use when native shell timing ( Read performance_report Generate performance report Use when native shell timing ( Read transfer_plugin-featured Get featured plugins from the store Use when native package install ( Read transfer_plugin-info Get detailed info about a plugin Use when native package install ( Read transfer_plugin-official Get official plugins from the store Use when native package install ( Read transfer_store-featured Get featured patterns from the store Use when native package install ( Read transfer_store-info Get detailed info about a pattern Use when native package install ( Read transfer_store-trending Get trending patterns from the store Use when native package install ( Read agent_health Compute an agent\ Read agent_list List every Ruflo-tracked agent in the registry with its type, model, status, and taskCount. Use when native Ta Read agent_metrics Get performance metrics for agents (V2 compatible). Deprecated: Use agent/status with includeMetrics instead. Read agent_status Get subagent status Read agentdb_graph-pathfinder Multi-algorithm native graph pathfinder (ADR-130 Phase 5). Use when agentdb_graph-query k-hop is not enough — Read agentdb_graph-query Unified graph traversal across the knowledge graph (ADR-130). Dispatches to the most capable backend: graph-no Read agentdb_pattern-search Search patterns via ReasoningBank controller with BM25+semantic hybrid Use when generic memory_ tools are wron Read aidefence_analyze Deep analysis of input for specific threat types with similar pattern search and mitigation recommendations. U Read aidefence_has_pii Check if input contains PII (emails, SSNs, API keys, passwords, etc.). Use when nothing native exists — Claude Read aidefence_is_safe Quick boolean check if input is safe. Fastest option for simple validation. Use when nothing native exists — C Read aidefence_scan Scan input text for AI manipulation threats (prompt injection, jailbreaks, PII). Returns threat assessment wit Read aidefence_stats Get AIDefence detection and learning statistics. Use when nothing native exists — Claude Code does not have a Read analyze_diff Analyze git diff for change risk assessment and classification Use when native Read analyze_diff-classify Classify git diff change type Use when native Read analyze_diff-reviewers Suggest reviewers for git diff changes Use when native Read analyze_diff-risk Quick risk assessment for git diff Use when native Read analyze_diff-stats Get quick statistics for git diff Use when native Read analyze_file-risk Assess risk for a specific file change Use when native Read autopilot_history Search past completion episodes by keyword. Requires AgentDB. Use when running long-horizon goals that should Read autopilot_learn Discover success patterns from past task completions. Requires AgentDB for full functionality. Use when runnin Read autopilot_log Retrieve the autopilot event log. Shows enable/disable events, re-engagements, completions. Use when running l Read autopilot_progress Detailed task progress broken down by source (team-tasks, swarm-tasks, file-checklist). Use when running long- Read autopilot_status Get autopilot state including enabled status, iteration count, task progress, and learning metrics. Use when r Read browser_cookie_use Fetch a vault handle for a host from the browser-cookies AgentDB namespace. Raw cookie values are NEVER return Read browser_get-text Get text content of an element Use when native WebFetch is wrong because you need real browser automation — JS Read browser_get-title Get the page title Use when native WebFetch is wrong because you need real browser automation — JS-heavy SPA s Read browser_get-url Get the current URL Use when native WebFetch is wrong because you need real browser automation — JS-heavy SPA Read browser_get-value Get value of an input element Use when native WebFetch is wrong because you need real browser automation — JS- Read browser_screenshot Capture screenshot of the page Use when native WebFetch is wrong because you need real browser automation — JS Read browser_session_replay Load a recorded session trajectory and return its steps so the caller can dispatch them through the 23 browser Read browser_session-list List active browser sessions Use when native WebFetch is wrong because you need real browser automation — JS-h Read browser_snapshot Get AI-optimized accessibility tree snapshot with element refs (@e1, @e2, etc.) Use when native WebFetch is wr Read cite Generate citation Read claims_list List all claims or filter by criteria Use when nothing native covers per-agent capability gating — Claude Code Read claims_load Get agent load information Use when nothing native covers per-agent capability gating — Claude Code agents hav Read config_get Get configuration value Use when native settings.json edits are wrong because the values need to be read by th Read config_list List configuration values Use when native settings.json edits are wrong because the values need to be read by Read coordination_load_balance Configure load balancing Use when native Task is wrong because the work crosses multiple agents that need to v Read coordination_metrics Get coordination metrics Use when native Task is wrong because the work crosses multiple agents that need to v Read cve_lookup Lookup CVE details Read daa_learning_status Get learning status for DAA agents Use when native Task is wrong because you need agents that adapt their cogn Read embeddings_compare Compare similarity between two texts Use when text similarity matters beyond keyword match — native Grep finds Read embeddings_hyperbolic Hyperbolic embedding operations (Poincaré ball) Use when text similarity matters beyond keyword match — native Read embeddings_neural Neural substrate operations (RuVector integration) Use when text similarity matters beyond keyword match — nat Read embeddings_rabitq_search Search via RaBitQ quantized index (fast Hamming scan). Returns candidate IDs for reranking. Use when text simi Read embeddings_rabitq_status Get RaBitQ quantized index status — availability, vector count, compression ratio Use when text similarity mat Read embeddings_search Semantic search across stored embeddings Use when text similarity matters beyond keyword match — native Grep f Read embeddings_status Get embeddings system status and configuration Use when text similarity matters beyond keyword match — native Read features_detect Detect runtime features (V2 compatible). Deprecated: Use system/info instead. Read federation_audit Query federation audit logs with optional compliance mode filtering Read federation_breaker_status ADR-097 Phase 4: per-peer circuit-breaker state snapshot. Returns each known peer with its lifecycle state (AC Read federation_evict ADR-097 Phase 4: operator-initiated evict for a peer. Marks the peer EVICTED so subsequent federation_send cal Read federation_join Join a federation by connecting to a remote peer endpoint Read federation_peers List all known federation peers with their trust levels and status Read federation_query Query federated memory from a remote peer (PII-gated) Read federation_reactivate ADR-097 Phase 4: operator-initiated reactivate for a SUSPENDED or EVICTED peer. Used after a health probe conf Read federation_status Get federation health status including active sessions, peers, and trust levels Read federation_trust View or review trust score details for a specific node Read federation_wg_status ADR-111 Phase 6: WG mesh state — per-peer trust level, mesh IP, suspended/evicted flags, and the AllowedIPs sl Read get-metrics Get collected metrics Read git-status Get current Git repository status Read github_metrics Get repository metrics and statistics Use when native Bash / file tools are wrong because this MCP tool expose Read github_repo_analyze Analyze a GitHub repository Use when native Bash / file tools are wrong because this MCP tool exposes Ruflo-sp Read grounded_query Query Google Search via Gemini 2.5 Flash grounding. Read gt_agents List Gas Town agents with optional role and rig filters Read gt_beads_ready List beads that are ready to work on (no unresolved dependencies) Read gt_beads_show Show detailed information about a specific bead including dependencies Read gt_convoy_status Check the status of one or all convoys including progress metrics Read gt_formula_list List available Gas Town formulas with optional type filter Read gt_wasm_match_pattern Find similar formulas or beads using HNSW pattern matching (150x-12500x faster) Read guidance Get instructions on how to use the available tool groups and services. Call this FIRST when unsure which tool Read guidance_capabilities List all capability areas with their tools, commands, agents, and skills. Use this to discover what Ruflo can Read guidance_discover Discover all available agents and skills from the .claude/ directory. Returns live filesystem data. Use when g Read guidance_quickref Quick reference card for common operations. Returns the most useful commands for a given domain. Use when gene Read guidance_recommend Given a task description, recommend which capability areas, tools, agents, and workflow to use. Use when gener Read guidance_workflow Get a recommended workflow template for a task type. Includes steps, agents, and topology. Use when generic Read hive-mind_status Get hive-mind status Use when native Task is wrong because you need queen-led collective intelligence — Byzant Read hooks_coverage-gaps List all coverage gaps with priority scoring and agent assignments Read hooks_coverage-suggest Suggest coverage improvements for a path (ruvector integration) Read hooks_explain Explain routing decision with full transparency Use when native Bash hooks (via Claude Code\ Read hooks_intelligence RuVector intelligence system status (shows REAL metrics from memory store) Use when native Bash hooks (via Cla Read hooks_intelligence_pattern-search Search patterns using REAL vector search (HNSW when available, brute-force fallback) Use when native Bash hook Read hooks_intelligence_stats Get RuVector intelligence layer statistics Use when native Bash hooks (via Claude Code\ Read hooks_intelligence_trajectory-step Record step in trajectory for reinforcement learning Use when native Bash hooks (via Claude Code\ Read hooks_list List all registered hooks Use when native Bash hooks (via Claude Code\ Read hooks_metrics View learning metrics dashboard Use when native Bash hooks (via Claude Code\ Read hooks_model-stats Get model routing statistics Use when native Bash hooks (via Claude Code\ Read hooks_session-end End current session, stop daemon, and persist state Use when native Bash hooks (via Claude Code\ Read hooks_worker-detect Detect worker triggers from user prompt (for UserPromptSubmit hook) Use when native Bash hooks (via Claude Cod Read hooks_worker-list List all 12 background workers with status and capabilities Use when native Bash hooks (via Claude Code\ Read hooks_worker-status Get status of a specific worker or all active workers Use when native Bash hooks (via Claude Code\ Read hyperbolic_hierarchy_compare Compare hierarchies using hyperbolic alignment. Computes structural and semantic similarity with node-level al Read hyperbolic_semantic_search Semantic search with hierarchical awareness. Supports nearest, subtree, ancestors, siblings, and cone search m Read hyperbolic_taxonomic_reason Taxonomic reasoning using hyperbolic entailment. Supports IS-A, subsumption, LCA, path, and similarity queries Read iot_custody_epoch Get custody epoch Read iot_device_list List all registered devices Read iot_device_pair Pair with a registered device to establish mutual trust Read iot_device_status Get device status and trust score Read iot_device_unpair Unpair from a registered device Read iot_firmware_rollout_list List firmware rollouts Read iot_firmware_rollout_status Get firmware rollout status Read iot_fleet_get Get fleet details Read iot_fleet_list List all fleets Read iot_fleet_status Get fleet overview Read iot_mesh_topology Get mesh network topology Read iot_store_query Query device vector store Read iot_store_status Get vector store health Read iot_telemetry_anomalies Detect anomalies in device telemetry readings Read iot_telemetry_baseline Compute or retrieve telemetry baseline for a device Read iot_witness_chain Get witness chain provenance Read iot_witness_verify Verify witness chain integrity for a device — detects epoch gaps and hash chain breaks Read list_files List all files in the virtual filesystem. No parameters needed. Returns file paths. Read managed_agent_events Fetch the full server-persisted event log of a managed cloud-agent session (user turns, agent thinking, tool_u Read managed_agent_list List managed cloud-agent sessions on this Anthropic org (id, status, title) — the CLOUD counterpart of wasm_ag Read managed_agent_status Get the lifecycle state of a managed cloud-agent session: idle/running/error, title, last error. Use when nati Read mcp_status Get MCP server status, including stdio mode detection Use when native Claude Code MCP status is wrong because Read memory_bridge_status Show Claude Code memory bridge status — AgentDB vectors, SONA learning, intelligence patterns, and connection Read memory_list Enumerate stored memory entries (optionally filtered by namespace/tags) without semantic search. Use when nati Read memory_retrieve Read back a value previously stored via memory_store, by exact (namespace, key) — lossless, includes metadata. Read memory_search Search semantic memory using HNSW Read memory_search_unified Search across both Claude Code memories and AgentDB entries using semantic vector similarity. Returns merged, Read neural_compress Compress neural model or embeddings Use when nothing native trains on your workflow — Claude Code has no learn Read neural_status Get neural system status Use when nothing native trains on your workflow — Claude Code has no learning loop. U Read pattern_search Search patterns (HNSW) Read pr_coherence_check Check coherence of vectors using Sheaf Laplacian energy (0=coherent, 1=contradictory) Read pr_consensus_verify Verify multi-agent consensus mathematically Read pr_quantum_topology Compute quantum topology features (Betti numbers, persistence) Read pr_spectral_analyze Analyze stability using spectral graph theory Read pre-browse Get recommendations before browser automation Read progress_check Get current V3 implementation progress percentage and metrics Use when native TodoWrite is wrong because you n Read progress_summary Get human-readable V3 implementation progress summary Use when native TodoWrite is wrong because you need cros Read progress_watch Get current watch status for progress monitoring Use when native TodoWrite is wrong because you need cross-ses Read read_file Read a file from the virtual filesystem (worker mock) Read report_vuln Generate vulnerability report Read retrieve-reasoning Retrieve similar reasoning patterns Read ruvector_health Check the health status of the RuVector connection and database. Read ruvector_index_stats Get statistics for vector indices. Read ruvector_metrics Get performance metrics and statistics. Read ruvector_search Search for similar vectors using HNSW or IVF indexing. Supports cosine, euclidean, and dot product distance me Read ruvllm_hnsw_route Route a query embedding to nearest patterns in HNSW index. Use when sending every prompt to the Anthropic API Read ruvllm_status Get ruvllm-wasm availability and initialization status. Use when sending every prompt to the Anthropic API is Read rvf_extract Extract RVF contents Read rvf_verify Verify RVF container Read scan-code Scan code for security vulnerabilities Read search Search your knowledge base for relevant information. Read session_info Get detailed session information Use when native conversation memory is wrong because you need durable cross-s Read session_list List saved sessions Use when native conversation memory is wrong because you need durable cross-session state Read swarm_health Check swarm health status with real state inspection Use when native Task tool is wrong because you need multi Read swarm_monitor Monitor swarm activity (V2 compatible). Deprecated: Use swarm/status with includeMetrics instead. Read swarm_status Get swarm status from persistent state Use when native Task tool is wrong because you need multi-agent coordin Read system_health Perform system health check Use when native Bash is wrong because you need Ruflo runtime metrics (HNSW index s Read system_info Get system information Use when native Bash is wrong because you need Ruflo runtime metrics (HNSW index size, Read system_metrics Get system metrics and performance data Use when native Bash is wrong because you need Ruflo runtime metrics ( Read system_status Get overall system status Use when native Bash is wrong because you need Ruflo runtime metrics (HNSW index siz Read task_list List all tasks Use when native TodoWrite is wrong because you need cross-session task persistence, agent assig Read task_results Retrieve results from completed tasks (V2 compatible). Deprecated: Use tasks/results instead. Read task_status Get task status Use when native TodoWrite is wrong because you need cross-session task persistence, agent assi Read task_summary Get a summary of all tasks by status Use when native TodoWrite is wrong because you need cross-session task pe Read teammate_delegate Delegate authority/permissions to another teammate Read teammate_discover_teams Discover existing teams in ~/.claude/teams/ Read teammate_find_optimal_path Find optimal message routing path between two teammates using BMSSP graph algorithms. Read teammate_get_status Get team and teammate status Read teammate_get_topology_stats Get topology statistics and optimization suggestions for a team. Read terminal_history Get command history for a terminal session Use when native Bash is wrong because you need a persistent termina Read terminal_list List all terminal sessions Use when native Bash is wrong because you need a persistent terminal session across Read todo_list List all pending tasks Read transfer_detect-pii Detect PII in content without redacting Use when native package install ( Read transfer_plugin-search Search the plugin store Use when native package install ( Read transfer_store-download Download a pattern from the store Use when native package install ( Read transfer_store-search Search the pattern store Use when native package install ( Read wasm_agent_files Get a WASM agent\ Read wasm_agent_is_stopped Check whether a WASM agent has reached its stop condition (max turns or explicit stop). Use when native Task i Read wasm_agent_list List all active WASM agents. Use when native Task is wrong because the workload needs sandboxed isolation — un Read wasm_agent_state Read the full internal state of a WASM agent (messages, turn count, config, stop status). Use when native Task Read wasm_agent_todos Get the structured todo list of a WASM agent as JSON. Use when native Task is wrong because the todo state liv Read wasm_agent_tools List the tools registered on a WASM agent sandbox. Use when native Task is wrong because the tool registry liv Read wasm_agent_turn_count Return the current turn count of a WASM agent. Use when native Task is wrong because turn-limit enforcement an Read wasm_gallery_active Return the ID of the currently active WASM gallery template. Use when native Bash is wrong because the active- Read wasm_gallery_categories Return all WASM gallery template categories with per-category template counts. Use when native Bash/ls is wron Read wasm_gallery_config Get the runtime configuration overrides applied to the active WASM gallery template. Use when native Read is w Read wasm_gallery_list List all available WASM agent gallery templates (Coder, Researcher, Tester, Reviewer, Security, Swarm). Use wh Read wasm_gallery_list_by_category List WASM gallery templates filtered to a specific category. Use when native Glob is wrong because gallery tem Read wasm_gallery_load_rvf Load a named gallery template as a base64-encoded RVF container. Use when native Read is wrong because RVF con Read wasm_gallery_search Search WASM agent gallery templates by query. Use when native Task is wrong because the workload needs sandbox Read web_research Search the web, fact-check claims, compare items, or conduct deep research. Actions: Read web_search Search the web using DuckDuckGo and return the top results (title, URL, snippet). Read witness_verify Verify chain integrity Read workflow_list List all workflows Use when native TodoWrite + sequential Bash is wrong because the work has a real dependency Read workflow_status Get workflow status Use when native TodoWrite + sequential Bash is wrong because the work has a real dependenc Read agentdb_context-synthesize Synthesize context from stored memories for a given query Use when generic memory_ tools are wrong because you Read agentdb_controllers List all AgentDB v3 controllers and their initialization status Use when generic memory_ tools are wrong becau Read agentdb_health Get AgentDB v3 controller health status including cache stats and attestation count Use when generic memory_ t Read agentdb_hierarchical-recall Recall from hierarchical memory with optional tier filter Use when generic memory_ tools are wrong because you Read config_export Export configuration to JSON Use when native settings.json edits are wrong because the values need to be read Read embeddings_generate Generate embeddings for text (Euclidean or hyperbolic) Use when text similarity matters beyond keyword match — Read hive-mind_memory Access hive shared memory Use when native Task is wrong because you need queen-led collective intelligence — B Read hooks_pre-edit Get context and agent suggestions before editing a file Use when native Bash hooks (via Claude Code\ Read hooks_route Get a 3-tier routing recommendation for a task: Tier 1 (Agent Booster, 0ms / $0 — for var-to-const, add-types, Read memory_stats Get memory storage statistics including HNSW index status Use when native Read/Write is wrong because you need Read wasm_gallery_export Export all custom WASM gallery templates as a JSON snapshot. Use when native Read/cat is wrong because custom
OTHER 2 tools
Other my-tool My custom tool Other noop Does nothing
EXPLORE

Related servers

Other MCP servers with similar tools — same risk classification, starter policies for each.

Yaver 1060 tools
paymentsadminautomation
Binance MCP Server 734 tools
paymentsadminautomation
GoHighLevel MCP Server 566 tools
paymentsadminautomation
ServiceTitan MCP Server 454 tools
paymentsadminautomation
0nmcp 407 tools
paymentsadminautomation
GoHighLevel MCP Server 406 tools
paymentsadminautomation
TheProtocol — Sovereign AI Agent Platform 380 tools
paymentsadminautomation
Aibtc 353 tools
paymentsadminautomation
FAQ

Questions about Claude Flow

Can an AI agent move money through the Claude Flow MCP server? +

Yes. The Claude Flow server exposes 3 financial tools including hooks_transfer, transfer_ipfs-resolve, federation_report_spend. Without a policy, an autonomous agent can call these with no spend caps, no rate limits, and no approval flow. PolicyLayer lets you block financial tools by default, require human approval, or set per-tool rate limits — enforced on every call.

Can an AI agent delete data through the Claude Flow MCP server? +

Yes. The Claude Flow server exposes 25 destructive tools including agentdb_batch, agentdb_causal-edge-delete, agentdb_causal-node-delete. These permanently remove resources with no undo. PolicyLayer blocks destructive tools by default so they never reach the upstream server.

How do I prevent bulk modifications through Claude Flow? +

The Claude Flow server has 91 write tools including gt_beads_dep, gt_convoy_track, hive-mind_leave. Set a rate limit in your policy -- for example, 10 calls per hour prevents an agent from making more than 10 modifications per hour. PolicyLayer enforces this at the gateway, before calls reach Claude Flow.

How many tools does the Claude Flow MCP server expose? +

454 tools across 5 categories: Destructive, Execute, Financial, Read, Write. 212 are read-only. 242 can modify, create, or delete data.

How do I enforce a policy on Claude Flow? +

Register the Claude Flow MCP server in PolicyLayer, apply the suggested rules above (adjust the limits to your use case), and point your AI client at the PolicyLayer proxy URL instead of the server directly. Your agents keep the same tools; PolicyLayer evaluates every call against policy before it executes. Nothing to install, live in minutes.

Enforce policy on every Claude Flow tool call.

Deterministic rules across all 454 Claude Flow tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON CLAUDE FLOW →

Instant setup, no code required.

454 Claude Flow tools catalogued and risk-classified — across an index of 43,000+ MCP servers.

// GET IN TOUCH

Have a question or want to learn more? Send us a message.

Message sent.

We'll get back to you soon.