// SCAN REPORT

Ralph Hero

54 tools. 24 can modify or destroy data without limits.

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

Last updated:

24 can modify or destroy data
30 read-only
54 tools total

Community server · catalogue entry verified 29/06/2026

How to control Ralph Hero ↓

TOOL MAP

What Ralph Hero exposes to your agents

Read (30) Write / Execute (22) Destructive / Financial (2)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS
Critical Risk

The most dangerous Ralph Hero tools

Destructive · Medium ralph_hero__remove_dependency Removing a dependency is an irreversible deletion of a relationship between GitHub issues. Destructive · High ralph_hero__sre__delete_pod Deleting a Kubernetes pod terminates the container and its workload, which cannot be undone without redeployment. Execute · High ralph_hero__sre__rollout_restart Rolling restarts of Kubernetes deployments are Execute-category operations because they trigger infrastructure changes through an external system. Execute · High ralph_hero__sre__scale This tool executes infrastructure operations on a live Kubernetes cluster.

24 of Ralph Hero's 54 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control Ralph Hero

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

Deny destructive operations
{
  "ralph_hero__remove_dependency": {
    "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
{
  "knowledge_record_outcome": {
    "limits": [
      {
        "counter": "knowledge_record_outcome_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations
{
  "knowledge_bridges": {
    "limits": [
      {
        "counter": "knowledge_bridges_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 Ralph Hero — 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 RALPH HERO →

Instant setup, no code required.

FULL CATALOGUE

All 54 Ralph Hero tools

DESTRUCTIVE 2 tools
Destructive ralph_hero__remove_dependency Remove a blocking dependency between two GitHub issues. Supports cross-repo: Destructive ralph_hero__sre__delete_pod Delete a single named pod in a Kubernetes namespace.
EXECUTE 3 tools
Execute ralph_hero__sre__drain Drain a Kubernetes node by evicting all non-daemonset pods. Execute ralph_hero__sre__rollout_restart Trigger a rolling restart of a Kubernetes deployment. Execute ralph_hero__sre__scale Scale a Kubernetes deployment to a specified replica count.
WRITE 19 tools
Write knowledge_record_outcome Record a pipeline outcome event (research, plan, phase, validation, etc.) Write knowledge_remember Persist a piece of text as a raw memory under ~/projects/thoughts/dream-memories/agent/YYYY/MM/DD/<source>-<ha Write ralph_hero__add_dependency Create a blocking dependency between two GitHub issues. Supports cross-repo: Write ralph_hero__add_sub_issue Create a parent/child (sub-issue) relationship between two GitHub issues. The parent issue becomes the contain Write ralph_hero__advance_issue Advance workflow state for related issues. Write ralph_hero__archive_items Archive or unarchive project items. Single-item mode: provide number or projectItemId (supports unarchive). Bu Write ralph_hero__batch_update Bulk-update project fields (workflow state, estimate, priority) across multiple issues in a single call. Uses Write ralph_hero__capture_snapshot Capture a single point-in-time snapshot of the project dashboard + metrics and append it to the partitioned JS Write ralph_hero__convert_draft_issue Convert a draft issue to a real repository issue. Requires the project item ID (PVTI_...) returned by create_d Write ralph_hero__create_comment Add a comment to a GitHub issue. Returns: commentId, issueNumber. Recovery: if issue not found, verify the iss Write ralph_hero__create_draft_issue Create a draft issue in the project (no repo required). Optionally set workflow state, priority, and estimate Write ralph_hero__create_issue Create a GitHub issue and add it to the project with optional field values. Returns: number, id, title, url, p Write ralph_hero__create_status_update Post a project-level status update with health designation. Visible in GitHub Projects UI header and panel. Re Write ralph_hero__create_views Copy views from a source GitHub Project V2 to a target project using the REST API. Reads view names, layouts, Write ralph_hero__decompose_feature Split a feature description into repo-specific issues using a named decomposition pattern from .ralph-repos.ym Write ralph_hero__save_issue Unified issue mutation: update any combination of issue properties (title, body, labels, assignees, open/close Write ralph_hero__setup_project Create a new GitHub Project V2 with Workflow State, Priority, Estimate, and optional Sprint iteration fields Write ralph_hero__sync_plan_graph Sync a plan document Write ralph_hero__update_draft_issue Update title and/or body of an existing draft issue. Requires the draft issue content node ID (DI_...), not th
READ 30 tools
Read knowledge_bridges Find bridge documents that connect different topic clusters. Uses betweenness centrality on an undirected grap Read knowledge_central Rank documents by PageRank (importance). Returns top documents by incoming citation weight. Isolated nodes (no Read knowledge_common Find documents that both docA and docB are connected to (shared neighbors). Returns enriched entries with rela Read knowledge_communities Detect document communities using Louvain algorithm. Returns clusters of related documents with modularity sco Read knowledge_community Fetch a single community by ID from Louvain detection. Returns the community Read knowledge_expert Return a domain-keyed memory bundle: curated wiki entries + recent reflections + prior outcomes for a named do Read knowledge_memory_stats Return counts of documents by memory_tier plus chunk percentiles and last-reflection timestamp. Used by the dr Read knowledge_paths Find all simple paths between two documents in the knowledge graph (max 20, direction-agnostic). Uses DFS with Read knowledge_query_outcomes Query outcome events with optional aggregation. Use to find patterns in pipeline history. Read knowledge_recall Role-aware retrieval over the knowledge base. Fans out one rerank-enabled hybrid.search() per memory tier in t Read knowledge_search Search the knowledge base by keyword, semantic similarity, and tags. Returns ranked documents. Read knowledge_subgraph Extract an N-hop neighborhood subgraph around a document. Returns deduplicated nodes and edges with distance f Read knowledge_think Query-time synthesis over the knowledge base. Retrieves the top-K grounding excerpts for a question, then asks Read knowledge_traverse Walk typed and untyped relationship edges from a document. Read ralph_hero__collate_debug Query Langfuse for error spans in a time window, normalize messages, group by signature, then either return th Read ralph_hero__debug_stats Aggregate debug log statistics. Reads JSONL logs and returns tool call counts, error rates, and average durati Read ralph_hero__detect_stream_positions Combined work-stream detection + per-stream pipeline position detection. Takes issue file ownership data and i Read ralph_hero__get_draft_issue Get the full content of one or more draft issues. Accepts DI_ (content node) or PVTI_ (project item) IDs — aut Read ralph_hero__get_issue Get a single GitHub issue with full context: properties, project field values, relationships (parent, sub-issu Read ralph_hero__get_project Get a GitHub Project V2 with all fields and their options Read ralph_hero__health_check Validate GitHub API connectivity, token permissions, repo access, project access, and required fields. Also su Read ralph_hero__list_dependencies List all blocking dependencies for a GitHub issue. Returns both Read ralph_hero__list_groups Discover all parent issues (groups) with sub-issues in a project. Fetches all project items (full project scan Read ralph_hero__list_issues List issues from a GitHub repository with optional filters. Fetches all project items (full project scan, no s Read ralph_hero__list_sub_issues List all sub-issues (children) of a parent GitHub issue, with completion summary. Use depth parameter (1-3) to Read ralph_hero__metrics_trends Read local snapshot JSONL and return 1d/7d/30d deltas plus sparklines for velocity, riskScore, wipTotal, leadT Read ralph_hero__next_actions Compute up to N deterministic Read ralph_hero__pipeline_dashboard Generate pipeline status dashboard with issue counts per workflow phase, health indicators, and formatted outp Read ralph_hero__project_hygiene Generate a project board hygiene report. Identifies archive candidates, stale items, orphaned backlog entries, Read ralph_hero__recent_activity Read structured events from the local ralph-hero activity log since a cursor. Used by /catch-up to synthesize
EXPLORE

Related servers

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

Yaver 1060 tools
adminautomation
Nodebench 824 tools
adminautomation
UnClick 813 tools
adminautomation
Binance MCP Server 734 tools
adminautomation
Fortimanager 584 tools
adminautomation
Crow 573 tools
adminautomation
Mcp 571 tools
adminautomation
GoHighLevel MCP Server 566 tools
adminautomation
FAQ

Questions about Ralph Hero

Can an AI agent delete data through the Ralph Hero MCP server? +

Yes. The Ralph Hero server exposes 2 destructive tools including ralph_hero__remove_dependency, ralph_hero__sre__delete_pod. 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 Ralph Hero? +

The Ralph Hero server has 19 write tools including knowledge_record_outcome, knowledge_remember, ralph_hero__add_dependency. 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 Ralph Hero.

How many tools does the Ralph Hero MCP server expose? +

54 tools across 4 categories: Destructive, Execute, Read, Write. 30 are read-only. 24 can modify, create, or delete data.

How do I enforce a policy on Ralph Hero? +

Register the Ralph Hero 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 Ralph Hero tool call.

Deterministic rules across all 54 Ralph Hero tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON RALPH HERO →

Instant setup, no code required.

54 Ralph Hero 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.