Menu

Individual skill

docsbot-evals

0upvotes

Use saved DocsBot Evals datasets, test reports, or instruction history only when the corresponding named tools are advertised. Otherwise draft manual test questions, review provided results, or improve current agent, Help Scout, legacy, or voice instructions through available bot tools.

skills/docsbot-administration/references/evals/SKILL.md

Read pinned source

Source commit fc97377adefd53e861c1d456aa1392e4a4e022a2

What was checked

Matches the observed repository HEAD
Yes
Scanner rules are current
No
Repository scan complete
No
Skill scan complete
No
Observed behavior
Not checked yet
Last checked
Nov 6, 2026

These are observations at the recorded commit, not a guarantee of safety. Review the source before running it.

Install this skill

Choose an agent in your Stack or copy a command below. Commands fetch the pinned commit and copy only this skill directory. Existing destinations are refused. Copying a command does not install anything.

An install command is unavailable: current rules, complete repository and skill scans, an eligible behavior level and a permitted license are required.

Repository context

Repository stars (not skill ratings)
62
Repository last updated
Nov 9, 2026
License
MIT

Observed agent integration: No specific integration evidence. Standard SKILL.md installation does not imply verified compatibility.

Files and excerpt

File metadata and permitted, redacted excerpts come from the recorded scan. Source links are pinned to that commit.

Browse files
Read excerpt

# Evals and instruction workflows

Use the current conversation and preserve the selected bot and channel. Before choosing a saved-set, test-run, or version-history workflow, call `search_tools` for that capability and inspect the returned names with `get_tool_schema`. These references describe optional capabilities; they do not establish that the Evals APIs or dashboard workflows have shipped. Never call a name missing from the currently advertised catalog. A dashboard handoff or bundled reference is not evidence that a tool is available.

If the required tools are absent, continue with a reviewable local question list, user-provided test results, normal conversation-history diagnosis, or current prompt edits through advertised `get_bot` and `update_bot`. Clearly say that saved Evals sets/runs or instruction-version history are unavailable through this connection; do not invent API routes, invoke legacy agents, or require those features for bot setup or response-quality work.

Keep the user oriented with the bot, channel, and any available question set or tests. Read only the resource needed for this task:

- Create or expand a question set, including from conversation history: [questions](references/questions.md).
- Explain test results or compare runs: [reports](references/reports.md).
- Improve or restore text or voice instructions: [instructions](references/instructions.md).

Call known named Admin MCP tools directly with their operation-specific `pathParams`, `query`, and/or `body`. Use `search_tools` and `get_tool_schema` only to inspect already advertised metadata when a name or schema is unclear. Do not use generic `execute` or an `operationId`. Do not call the legacy standalone eval-set-draft, eval-set-agent, run-analyze, prompt, or prompt-debug agents. Tool availability comes from the currently advertised catalog and permissions are checked on each action; if an operation is absent, prepare a concrete proposal and explain the missing capability without claiming it was saved.
Back to top