Scores an agent's trace 0-100 with concrete fixes. No content needed: send step types only.
Maintained by@TGS-checker(verified)Also a maintainer?
Call this server's tools from your browser. Nothing to install.
Requests go through mcptry because browsers can't call most MCP servers directly. Header values are sent to the server for that request only and are never stored or logged.
Tool list from our last check. Connect live to refresh it.
Get an independent, deterministic reliability check of a multi-step task you (an agent) just ran. Returns a 0-100 score, the issues found, and concrete fixes to apply on your next attempt. WHEN TO CALL: after you finish (or give up on) a multi-step task that used tools, before you report "done" to the user, or after a failure you want to learn from. Free, no API key, fast and offline (keyword heuristics, no LLM). Call get_trace_schema first if unsure of the format. INPUT (all plain strings): - goal: the original task, stated specifically (what success looks like). Vague goals under 20 chars are penalised. - trace: your chronological steps, one per line, e.g. "1. Called search_flights(AMS->LIS) -> returned 3 results" / "2. Observed: ..." . Write what each tool RETURNED ("observed", "returned", "result:"), any error and the retry/fallback you used, and a final "Verified: ..." line. - outcome: the concrete final result you claim, and how it meets the goal. - agent_name: optional identifier for yourself (logging only). RETURNS JSON: score (int 0-100), dimensions {goal_clarity, observability, error_recovery, goal_alignment, efficiency}, issues (list), suggestions (list, max 5), summary, version. THE LOOP: run task -> evaluate_trace -> apply the suggestions to your process -> re-run -> evaluate_trace again and compare scores/dimensions. Scores below ~60 usually mean missing observations, unrecovered errors, repeated steps (loops), success claimed without evidence, or no verification step.
The original task, stated specifically, including what success looks like (e.g. 'Find the cheapest direct AMS->LIS flight on 12 March and report price and flight number'). Goals under 20 characters are penalised.
Your chronological steps, one per line (numbered). For each tool call write what you called and what it returned/what you observed ('Observed: ...', '-> returned ...'). Record every error AND the retry/fallback you used. End with a 'Verified: ...' line that checks the outcome against the goal.
The concrete final result you claim (at least 15 characters) and how it satisfies the goal. Back success claims with evidence (IDs, values, 'tests passed', 'status 200').
Optional identifier for yourself or your agent (e.g. 'my-travel-agent'). Used for logging and usage counts only; it does not affect the score.
Pick your client. Values in capitals, like YOUR_TOKEN, are yours to fill in.
Run in a terminal
claude mcp add --transport http tracecheck https://tracecheck-mcp.onrender.com/mcp
Tool list last changed Nov 22, 2026.
Get an independent, deterministic reliability check of a multi-step task you (an agent) just ran. Returns a 0-100 score, the issues found, and concrete fixes to apply on your next attempt. WHEN TO CALL: after you finish (or give up on) a multi-step task that used tools, before you report "done" to the user, or after a failure you want to learn from. Free, no API key, fast and offline (keyword heuristics, no LLM). Call get_trace_schema first if unsure of the format. INPUT (all plain strings): - goal: the original task, stated specifically (what success looks like). Vague goals under 20 chars are penalised. - trace: your chronological steps, one per line, e.g. "1. Called search_flights(AMS->LIS) -> returned 3 results" / "2. Observed: ..." . Write what each tool RETURNED ("observed", "returned", "result:"), any error and the retry/fallback you used, and a final "Verified: ..." line. - outcome: the concrete final result you claim, and how it meets the goal. - agent_name: optional identifier for yourself (logging only). RETURNS JSON: score (int 0-100), dimensions {goal_clarity, observability, error_recovery, goal_alignment, efficiency}, issues (list), suggestions (list, max 5), summary, version. THE LOOP: run task -> evaluate_trace -> apply the suggestions to your process -> re-run -> evaluate_trace again and compare scores/dimensions. Scores below ~60 usually mean missing observations, unrecovered errors, repeated steps (loops), success claimed without evidence, or no verification step.
| Argument | Type | Description |
|---|---|---|
| goal* | string | The original task, stated specifically, including what success looks like (e.g. 'Find the cheapest direct AMS->LIS flight on 12 March and report price and flight number'). Goals under 20 characters are penalised. |
| trace* | string | Your chronological steps, one per line (numbered). For each tool call write what you called and what it returned/what you observed ('Observed: ...', '-> returned ...'). Record every error AND the retry/fallback you used. End with a 'Verified: ...' line that checks the outcome against the goal. |
| outcome* | string | The concrete final result you claim (at least 15 characters) and how it satisfies the goal. Back success claims with evidence (IDs, values, 'tests passed', 'status 200'). |
| agent_name | string | Optional identifier for yourself or your agent (e.g. 'my-travel-agent'). Used for logging and usage counts only; it does not affect the score. |
{
"type": "object",
"title": "evaluate_traceArguments",
"required": [
"goal",
"trace",
"outcome"
],
"properties": {
"goal": {
"type": "string",
"title": "Goal",
"description": "The original task, stated specifically, including what success looks like (e.g. 'Find the cheapest direct AMS->LIS flight on 12 March and report price and flight number'). Goals under 20 characters are penalised."
},
"trace": {
"type": "string",
"title": "Trace",
"description": "Your chronological steps, one per line (numbered). For each tool call write what you called and what it returned/what you observed ('Observed: ...', '-> returned ...'). Record every error AND the retry/fallback you used. End with a 'Verified: ...' line that checks the outcome against the goal."
},
"outcome": {
"type": "string",
"title": "Outcome",
"description": "The concrete final result you claim (at least 15 characters) and how it satisfies the goal. Back success claims with evidence (IDs, values, 'tests passed', 'status 200')."
},
"agent_name": {
"type": "string",
"title": "Agent Name",
"default": "unknown",
"description": "Optional identifier for yourself or your agent (e.g. 'my-travel-agent'). Used for logging and usage counts only; it does not affect the score."
}
}
}Return the exact input format TraceCheck's evaluate_trace expects, the scored dimensions and weights, and a complete example call you can copy. Call this once before your first evaluate_trace. No arguments, no side effects.
No arguments.
{
"type": "object",
"title": "get_trace_schemaArguments",
"properties": {}
}We read every tool's name, description and inputs the way the model will, looking for instructions hidden from you. A clean result means no warning signs, not a guarantee. Scanned Nov 22, 2026.
What changed in this server's tools since we started watching, newest first. Each change is scanned again.
{
"type": "object",
"title": "evaluate_traceArguments",
"required": [
"goal",
"trace",
"outcome"
],
"properties": {
"goal": {
"type": "string",
"title": "Goal"
},
"trace": {
"type": "string",
"title": "Trace"
},
"outcome": {
"type": "string",
"title": "Outcome"
},
"agent_name": {
"type": "string",
"title": "Agent Name",
"default": "unknown"
}
}
}{
"type": "object",
"title": "evaluate_traceArguments",
"required": [
"goal",
"trace",
"outcome"
],
"properties": {
"goal": {
"type": "string",
"title": "Goal",
"description": "The original task, stated specifically, including what success looks like (e.g. 'Find the cheapest direct AMS->LIS flight on 12 March and report price and flight number'). Goals under 20 characters are penalised."
},
"trace": {
"type": "string",
"title": "Trace",
"description": "Your chronological steps, one per line (numbered). For each tool call write what you called and what it returned/what you observed ('Observed: ...', '-> returned ...'). Record every error AND the retry/fallback you used. End with a 'Verified: ...' line that checks the outcome against the goal."
},
"outcome": {
"type": "string",
"title": "Outcome",
"description": "The concrete final result you claim (at least 15 characters) and how it satisfies the goal. Back success claims with evidence (IDs, values, 'tests passed', 'status 200')."
},
"agent_name": {
"type": "string",
"title": "Agent Name",
"default": "unknown",
"description": "Optional identifier for yourself or your agent (e.g. 'my-travel-agent'). Used for logging and usage counts only; it does not affect the score."
}
}
}