// SCAN REPORT

Codecks

38 tools. 19 can modify or destroy data without limits.

1 destructive tool with no built-in limits. Policy required.

Last updated: 3 Jul 2026

19 can modify or destroy data

19 read-only

38 tools total

Community server · catalogue entry verified 03/07/2026

How to control Codecks ↓

TOOL MAP

What Codecks exposes to your agents

Read (19) Write / Execute (18) Destructive / Financial (1)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

Critical Risk

The most dangerous Codecks tools

Destructive · High delete_card This tool irreversibly removes project management data (a card) from the system with no undo capability. Write · Low remove_from_hand Removing a card from a user's 'hand' in Codecks is a reversible write operation - the card still exists and can be re-added. Write · Medium planning_init This tool creates new files in the project root directory. Write · Medium add_to_hand The tool creates or modifies an assignment relationship between cards and a user, which is a reversible data modification.

19 of Codecks's 38 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control Codecks

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

Deny destructive operations

{
  "delete_card": {
    "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

{
  "add_to_hand": {
    "limits": [
      {
        "counter": "add_to_hand_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "get_account": {
    "limits": [
      {
        "counter": "get_account_per_minute",
        "window": "minute",
        "max": 60,
        "scope": "grant"
      }
    ]
  }
}

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

Create a free account and register Codecks — nothing to install.
Add these rules — paste them, or build them visually. Tune the limits to your setup.
Point your MCP client (Claude, Cursor, anything) at your gateway URL.

ENFORCE POLICY ON CODECKS →

Instant setup, no code required.

FULL CATALOGUE

All 38 Codecks tools

DESTRUCTIVE 1 tools

Destructive delete_card Permanently delete a card. Cannot be undone — use archive_card if reversibility needed.

WRITE 18 tools

Write add_to_hand Add cards to the user Write archive_card Archive a card (reversible). Write close_comment Close (resolve) a comment thread. Write create_card Create a new card. Set deck/project to place it. Use parent to nest as sub-card. Write create_comment Start a new comment thread on a card. Write mark_done Mark cards as done. Write mark_started Mark cards as started. Write planning_init Create lean planning files (task_plan.md, findings.md, progress.md) in project root. No auth needed. Write planning_update Update planning files mechanically. No auth needed. Operations: goal, advance, phase_status, error, decision, Write remove_from_hand Remove cards from the user Write reopen_comment Reopen a closed comment thread. Write reply_comment Reply to an existing comment thread. Write save_cli_feedback Save a CLI feedback item for the development team. No auth needed. Write save_workflow_preferences Save observed workflow patterns from current session. No auth needed. Write scaffold_feature Create a Hero card with Code/Design/Art/Audio sub-cards. Write split_features Batch-split unsplit feature cards into lane sub-cards. Use dry_run=true to preview. Write unarchive_card Restore an archived card. Write update_cards Update card properties. Doc cards: only owner/tags/milestone/deck/title/content/hero.

READ 19 tools

Read get_account Get current account info (name, id, email, role). Read get_card Get full card details (content, checklist, sub-cards, conversations, hand status). Read get_cli_feedback Read saved CLI feedback items. Optionally filter by category. No auth needed. Read get_lane_registry Get the local lane (deck category) definitions and metadata. No auth needed. Read get_pm_playbook Get PM session methodology guide. No auth needed. Read get_tag_registry Get the local tag taxonomy (definitions, hero tags, lane-tag mappings). No auth needed. Read get_workflow_preferences Load user workflow preferences from past sessions. No auth needed. Read list_activity Show recent activity feed. Read list_cards List cards. Filters combine with AND. Read list_conversations List all comment threads on a card with messages and thread IDs. Read list_decks List all decks. Set include_card_counts=True for per-deck counts. Read list_hand List cards in the user Read list_milestones List all milestones with card counts. Read list_projects List all projects with deck info. Read planning_measure Track token usage of planning files over time. No auth needed. Operations: snapshot, report, compare_templates Read planning_status Get compact planning status: goal, phases, decisions, errors, token count. No auth needed. Read pm_focus PM focus dashboard: blocked, stale, unassigned, and suggested next cards. Read standup Daily standup summary: recently done, in-progress, blocked, and hand.

EXPLORE

Related servers

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

FAQ

Questions about Codecks

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

Yes. The Codecks server exposes 1 destructive tools including delete_card. 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 Codecks? +

The Codecks server has 18 write tools including add_to_hand, archive_card, close_comment. 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 Codecks.

How many tools does the Codecks MCP server expose? +

38 tools across 3 categories: Destructive, Read, Write. 19 are read-only. 19 can modify, create, or delete data.

How do I enforce a policy on Codecks? +

Register the Codecks 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 Codecks tool call.

Deterministic rules across all 38 Codecks tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON CODECKS →

Instant setup, no code required.

38 Codecks tools catalogued and risk-classified — across an index of 43,000+ MCP servers.

// WHERE THIS COMES FROM

These policies come from Codecks's registry record.

The record behind this page: verified identity, auth posture, risk grade, every tool classified, recommended policy — re-checked continuously.

Teams ship this data inside their own products. See what a licence covers →