Track file edits and test results to detect and prevent death spirals in AI coding sessions. Check safety before editing.
MIT
381,317 tools. Last updated 2026-08-03 01:39
"Loop" matching MCP tools:
- Create a Microsoft Loop page and populate it with Markdown. Automates the paste-to-blocks conversion, turning your content into formatted Loop blocks.MIT
- Read any Microsoft Loop page as Markdown by providing its URL. Opens the page in your authenticated Loop session and extracts the rendered content.MIT
- Register a custom loop to the local MCP by providing full loop text; it is hashed, split into phases, and persisted for streaming.
- Start phase-gated streaming of a bundled or custom loop, activating the supervisor lane to return the initial section while the full loop runs in the background.
- Activate a supervisor lane for phase-gated streaming of a bundled or custom local loop. Returns only section 0; the full loop stays in the supervisor.
Matching MCP Servers
- AlicenseAqualityCmaintenanceEnables reading, listing, creating, and editing Microsoft Loop pages as Markdown through browser automation.Last updated4MIT
- Alicense-qualityBmaintenanceDingDawg Loop Protocol (DDLP) — safe scheduled AI agents with governance gates. Every loop execution is verified, receipted, and fail-closed. MCP-native, works with CrewAI, LangGraph, Claude Code, Cursor.Last updated64MIT
Matching MCP Connectors
L402 MCP: 5 paid BTC/Lightning tools + fiat credits, 10-25 sats/call.
Cloudflare Workers MCP server: agent-loop-detector
- Retrieve autonomous loop session metrics including iterations completed, cost, errors, and safeguard status. Use to inspect runtime outcomes from an active or completed session.MIT
- Completes the current loop task, records session metrics, and returns the next task to continue the cycle.MIT
- Configure autonomous build loop safeguards by setting max iterations, cost cap, error threshold, and approval policy before starting a session.MIT
- Lint any loop design against an anti-pattern rubric to reveal missing verifier, stop condition, budget, and other critical checks. Returns a score, fixes, and related loops.MIT
- Check AI agent loop detection metrics: call count, unique signatures, max identical calls, and loop risk (LOW/MEDIUM/HIGH). Identifies high-risk loops that may trip a circuit breaker.MIT
- Get a snapshot of the work loop catalog: total loops, categories, featured items, last update, and source. This helps assess available bounded protocols for coding tasks.MIT
- Initiate a guarded task loop for AI agents to implement repository changes with automatic verification, audit, and background continuation.MIT
- Scan the full loop library as a digest with id, category, use_when, and verifier strength, then choose the best fit for your context.MIT
- Check the status of a loop run by run ID to see iterations, cost, tokens, flagged calls, and completion, helping agents decide whether to continue iterating.MIT
- Summarize loop run records from saved files or run-store directories for auditing AI agent actions and costs.Apache 2.0
- Scaffold a new loop spec from a stated bottleneck when no catalog loop fits. The draft includes every required field, an MVL sanity check, and nearest existing loops for structure.MIT
- Discover loop categories with live counts and direct filter URLs for targeted browsing of the AI Loop Library.MIT
- List pending Human-in-the-Loop verifications for a group to see checkpoints and votes awaiting human approval.MIT
- Check the current budget and cost state of a Martin loop record to prevent runaway loops and control spending.Apache 2.0