CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

search

Answer questions from the user's own files and chat/mail archives through the AutoRAG MCP tools. Use when the user asks to find, summarize, compare, or verify something in local documents (PDF, DOCX, PPTX, XLSX, HWP/HWPX, Markdown, email) or in connected Slack, Discord, Telegram, WhatsApp, KakaoTalk, email, Notion, or Obsidian archives. Not for web search and not for first-time setup (use the setup skill).

Quality

96%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

96%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is a dense, highly operational guide: exact tool invocations with JSON arguments, a sequenced workflow with retry and error-recovery loops, and explicit safety boundaries. Its only structural weakness is that all operational detail is inlined in a single file with no progressive-disclosure layer for the longer reference material.

DimensionReasoningScore

Conciseness

The body contains zero explanations of concepts Claude already knows; every sentence carries tool-specific operational detail (e.g., "the `<id>` is lower-cased with every run of non-alphanumerics collapsed to `_`"). Lean and efficient throughout, matching the every-token-earns-its-place anchor rather than the minor-trimming anchor at 4.

5 / 5

Actionability

Fully actionable with exact tool names and argument objects: `autorag.search` with `{"query": "...", "topK": 8}`, `autorag.refresh` with `{"methods": ["parsed", "minsync"]}` and `{"force": true}`, plus named result-envelope fields ("`number`, `source`, `method`, `score`, `metadata`, and `content`"). As an instruction-only skill, the guidance is copy-paste-ready equivalent.

5 / 5

Workflow Clarity

A clear 6-step sequence with feedback loops ("Search with 2-3 differently worded queries when the first results do not contain the answer") and a dedicated error-recovery section covering `index-not-ready`, `stale: true`, `unsearched`, and `datasource-*` diagnostics. The destructive/batch cap does not apply: refresh "writes only AutoRAG's own index directories" and duplicates is read-only with "never move or delete files on your own".

5 / 5

Progressive Disclosure

Good structure with clear sections (Workflow, Reading the result envelope, Refresh and duplicates, Safety) and a clearly signaled pointer to the `setup` skill. Not 5 because at ~105 lines — above the under-50-line simple-skill exception — with no bundle files, sections like the envelope diagnostics and the convert/duplicates detail could arguably be split into reference files.

4 / 5

Total

19

/

20

Passed

Description

95%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is exemplary: concrete actions, comprehensive natural trigger terms spanning file formats and chat platforms, explicit use-when guidance, and clear negative boundaries. The only minor gap is that secondary capabilities described in the body (refresh, duplicates, conversion) are absent from the description.

DimensionReasoningScore

Specificity

Quotes: "Answer questions from the user's own files and chat/mail archives through the AutoRAG MCP tools" and "find, summarize, compare, or verify" — several concrete actions with named formats and sources. Not 5 because body capabilities (refresh, duplicates, convert) are not surfaced, leaving minor coverage gaps; clearly above 3 with more than 1-2 concrete actions.

4 / 5

Completeness

Explicitly answers both: what ("Answer questions from the user's own files and chat/mail archives through the AutoRAG MCP tools") and when ("Use when the user asks to find, summarize, compare, or verify something in local documents..."), plus negative boundaries ("Not for web search and not for first-time setup").

5 / 5

Trigger Term Quality

Quotes: "PDF, DOCX, PPTX, XLSX, HWP/HWPX, Markdown, email", "Slack, Discord, Telegram, WhatsApp, KakaoTalk, email, Notion, or Obsidian", and natural verbs "find, summarize, compare, or verify" — comprehensive natural phrasing including file extensions and app names users would actually say.

5 / 5

Distinctiveness Conflict Risk

Clear niche — searching the user's own local documents and chat/mail archives — with explicit exclusions ("Not for web search and not for first-time setup (use the setup skill)") that minimize conflict with adjacent skills.

5 / 5

Total

19

/

20

Passed

Validation

100%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository

Marker-Inc-Korea/AutoRAG

Path

plugins/hermes/skills/search/SKILL.md

Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.