Reduces AI Agent token usage by 40% via three-stage SOP workflow.
Glama couldn't complete the latest health check. If this server requires authentication, missing or expired test credentials may be the cause. A test profile lets Glama authenticate for health checks and discover tools; it is separate from your personal connections.
If you are the author, claim ownership, then add or update a test profile under Admin → Test Profile.
Scored across 4 tools
Each tool serves a distinct phase of the protocol: start_task initiates, report_step_result logs progress, finalize_and_report concludes, and get_network_stats provides external context. There is no overlap or ambiguity between their purposes.
All tool names follow a consistent verb_noun pattern in snake_case: start_task, report_step_result, finalize_and_report, get_network_stats. The naming style is uniform and predictable.
With exactly four tools, the server is tightly scoped to its protocol lifecycle. Each tool is necessary and the count is neither excessive nor sparse for the intended workflow.
The tool set covers the complete lifecycle of the Huangting Protocol: start, per-step reporting, and finalization. The addition of get_network_stats provides useful monitoring without leaving any obvious operational gap.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
start_task1 field changedInput schema / properties / task_typeAdded value: +{
+ "default": "default",
+ "description": "Task category for accurate baseline token modeling. Determines context_multiplier, expected steps, and output length. Choose the closest match: complex_research (deep research, 15 steps, ×4 context), code_generation (8 steps, ×2.5), multi_agent_coordination (20 steps, ×5), relationship_analysis (6 steps, ×2), optimization (5 steps, ×1.8), writing (4 steps, ×1.5), data_analysis (8 steps, ×2.5), default (8 steps, ×2.5).",
+ "enum": [
+ "complex_research",
+ "code_generation",
+ "multi_agent_coordination",
+ "relationship_analysis",
+ "optimization",
+ "writing",
+ "data_analysis",
+ "default"
+ ],
+ "type": "string"
+}create_optimization_contextfinalize_and_reportreport_optimization_resultreport_step_resultstart_taskreport_optimization_result1 field changedInput schema / properties / task_typeAdded value: +{
+ "description": "Optional: categorize the task type for analytics (e.g. 'complex_research', 'code_generation', 'multi_agent_coordination'). Defaults to 'optimization'.",
+ "type": "string"
+}create_optimization_contextget_optimization_strategyget_protocol_conceptreport_optimization_result7 fields changedInput schema / properties / actual_tokens_usedAdded value: +{
+ "description": "Actual total tokens consumed during the optimized task run",
+ "minimum": 1,
+ "type": "integer"
+}Input schema / properties / baseline_tokensAdded value: +{
+ "description": "Baseline token estimate from create_optimization_context response",
+ "minimum": 1,
+ "type": "integer"
+}Input schema / properties / context_idAdded value: +{
+ "description": "The context_id returned by create_optimization_context",
+ "type": "string"
+}Input schema / properties / task_typeRemoved value: -{
- "description": "Type of task that was optimized",
- "enum": [
- "complex_research",
- "code_generation",
- "multi_agent_coordination"
- ],
- "type": "string"
-}Input schema / properties / tokens_baselineRemoved value: -{
- "description": "Baseline token count without optimization. Measure BEFORE applying Huangting optimization.",
- "minimum": 1,
- "type": "integer"
-}Input schema / properties / tokens_savedRemoved value: -{
- "description": "Number of tokens saved compared to unoptimized baseline. Must be <= tokens_baseline.",
- "minimum": 0,
- "type": "integer"
-}Input schema / requiredPrevious value: -[
- "agent_id",
- "task_type",
- "tokens_saved",
- "tokens_baseline"
-]New value: +[
+ "agent_id",
+ "actual_tokens_used",
+ "baseline_tokens"
+]report_optimization_result5 fields changedInput schema / properties / tokens_baseline / defaultRemoved value: -0Input schema / properties / tokens_baseline / descriptionPrevious value: -"Baseline token count without optimization (for ratio calculation)"New value: +"Baseline token count without optimization. Measure BEFORE applying Huangting optimization."Input schema / properties / tokens_baseline / minimumPrevious value: -0New value: +1Input schema / properties / tokens_saved / descriptionPrevious value: -"Number of tokens saved compared to unoptimized baseline"New value: +"Number of tokens saved compared to unoptimized baseline. Must be <= tokens_baseline."Input schema / requiredPrevious value: -[
- "agent_id",
- "task_type",
- "tokens_saved"
-]New value: +[
+ "agent_id",
+ "task_type",
+ "tokens_saved",
+ "tokens_baseline"
+]get_network_statsget_optimization_strategyget_protocol_conceptreport_optimization_resultShared distillation cache for AI agents — every fetch ~73-89% fewer tokens via a shared cache.
SaaS intelligence for AI agents. 5 unified tools cover 1,000+ services with 91-96% token savings.
AI agent skills marketplace — token-efficient skill search & execution
Memory that reasons: continual learning for stateful agents. Better context, fewer tokens.
Add one secure layer between your agents and this server.
