MCP protocol adapters for tracegc
Wellknown found it in public sources; nobody has proven control of it yet. Claiming takes one click if the repository is under your GitHub account, or a small file on your domain otherwise. Verified owners get the badge, 15-minute checks, status alerts, edits that outrank crawled data, and a ranking boost.
Agents can do it too: POST https://wellknown.network/api/v1/claims with {"agent":"tracegc-mcp","method":"well_known_file"} — machine-readable steps at claim.json, guide at /docs/claim.
Everything here was measured by our prober or read from a registry. Nothing is self-reported.
Attributed to the source that supplied each field. Treated as claims, not facts.
# tracegc-mcp Model Context Protocol (MCP) server integration for TraceGC context compaction. ## Usage Example `tracegc-mcp` exposes three core MCP tools backed by TraceGC's deterministic compaction pipeline: - `add_event(event: dict, session_id: str)` — Appends a structured event to an active session trace. - `compact(session_id: str)` — Runs dead-branch sweeping, override pruning, deduplication, and cycle collapse for a session, returning the compacted prompt and metrics. - `get_receipt(session_id: str, node_id: str)` — Retrieves full, unpruned event payload receipts for any pruned node. ```python from tracegc_mcp import TraceGCMCPServer, add_event, compact, get_receipt # 1. Initialize server instance server = TraceGCMCPServer() # 2. Add events to a session server.add_event({ "id": "e1", "type": "set_var", "timestamp": 100, "key": "database", "value": "postgres" }, session_id="sess_100") server.add_event({ "id": "e2", "type": "set_var", "timestamp": 200, "key": "database", "value": "sqlite" }, session_id="sess_100") # 3. Compact the session trace result = server.compact(session_id="sess_100") print(result["prompt"]) # Output: "database = sqlite" # 4. Retrieve receipt for the pruned event receipt = server.get_receipt(session_id="sess_100", node_id="e1") print(receipt["receipt"]["value"]) # Output: "postgres" ``` Module-level tool helpers (`add_event`, `compact`, `get_receipt`) are also provided for direct tool dispatching.
Mapped onto the structured taxonomy from declared text and observed tool names. Confidence shown for derived entries.
Every source is kept verbatim. Field changes are logged as events.