truesight-workflows
npx machina-cli add skill Goodeye-Labs/truesight-mcp-skills/truesight-workflows --openclawTruesight Workflows
Use this skill as the routing entrypoint across all Truesight MCP skills.
Role and scope
This skill is a router. It decides intent and routes to exactly one skill path.
Do not execute deep workflow steps here unless the user already asked for a very specific action and no further routing is needed.
Routing map
- Build custom live eval from scratch ->
create-evaluation - Evaluate one or more traces with an existing live eval ->
evaluate-trace - Analyze failure modes in dataset traces ->
error-analysis - Judge flagged items and add labeled outputs back to dataset ->
review-and-promote-traces - Start quickly from pre-built template ->
bootstrap-template-evaluation - Audit current eval setup and maturity ->
eval-audit - Build custom review web interface ->
build-review-interface
Interactive Q&A protocol (mandatory)
When user intent is unclear, ask one question at a time and use lettered options whenever practical so the assistant can use structured selection widgets.
Question format:
Which workflow do you want to run first?
A) Evaluate traces with an existing live eval
B) Run error analysis on a dataset
C) Review and promote flagged traces
D) Bootstrap from a template
E) Create a new evaluation from scratch
F) Audit my eval setup
G) Build a custom review interface
Rules:
- Ask exactly one routing question per message.
- Use one follow-up question only if the answer is still ambiguous.
- After routing is clear, hand off immediately to the target skill.
Guardrails
- If user asks for
create-evaluation, do not decompose it into smaller skills. - Keep guidance scoped to currently available Truesight MCP tools.
- If user asks for functionality outside current MCP capabilities, state the gap clearly and offer the closest supported workflow.
Source
git clone https://github.com/Goodeye-Labs/truesight-mcp-skills/blob/main/skills/truesight-workflows/SKILL.mdView on GitHub Overview
Truesight Workflows serves as the routing entrypoint for all MCP skills. It decides user intent and routes to exactly one skill path (e.g., create-evaluation, evaluate-trace, error-analysis, review-and-promote-traces, bootstrap-template-evaluation, eval-audit, or build-review-interface) and avoids deep workflow steps unless a very specific action is requested.
How This Skill Works
It acts as a router that analyzes user intent and dispatches to a single target skill path. It does not execute deep workflow steps here unless the user asks for a very specific action and no further routing is needed. When the intent is unclear, it follows the mandatory interactive Q&A protocol and asks exactly one routing question with lettered options, then hands off to the chosen skill.
When to Use It
- Build custom live eval from scratch
- Evaluate traces with an existing live eval
- Analyze failure modes in dataset traces
- Judge flagged items and add labeled outputs back to dataset
- Start quickly from pre-built template
- Audit current eval setup and maturity
- Build a custom review interface
Quick Start
- Step 1: Identify user intent and map to a MCP skill
- Step 2: Present the routing question with A) Evaluate traces, B) Error analysis, C) Review, D) Bootstrap template, E) Create evaluation, F) Audit, G) Build review interface
- Step 3: Route to the chosen skill and hand off, with no further steps executed here
Best Practices
- Refer to the routing map to ensure the correct target path
- Ask exactly one routing question per message
- Use the lettered A-G options to enable structured selection
- Do not decompose to deeper steps in this skill unless a specific action is requested
- Hand off immediately to the target skill after routing
Example Use Cases
- User asks to evaluate traces with a live eval, routing to evaluate-trace
- User requests a quick start from a template, routing to bootstrap-template-evaluation
- User intent is ambiguous; the assistant asks one routing question and then routes
- User wants to audit the current eval setup, routing to eval-audit
- User aims to build a custom review interface, routing to build-review-interface
