CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

setup

One-pass setup for the AutoRAG plugin - pick folders, write the config, auto-connect the Slack, Discord, Telegram, WhatsApp, KakaoTalk, and email archives already on this machine, build the first index, and prove a search works. Use when autorag.status or a search says index-not-ready, no folders are configured, a datasource is missing, search is keyword-only, or the autorag tools are absent.

Quality

95%

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.

An excellent operational skill body: fully executable commands, measured timings and failure modes instead of generic advice, and a tightly sequenced workflow with real validation loops. The sole structural flaw is the reference to a docs/datasource-skills.md file that does not exist in the bundle.

Suggestions

Add the referenced docs/datasource-skills.md to the bundle (or inline the remaining datasource entry shapes into the table) so the KakaoTalk/email row's reference resolves.

Consider moving the long Troubleshooting section's rare failure cases (e.g., the hermes pm installer PATH issue) into a reference file to slim the main body further.

DimensionReasoningScore

Conciseness

The body is lean and operational throughout; every "Measured:" aside carries non-obvious, decision-relevant data (the macOS TCC read hang, the 2.5-minute first index, the MCP server ignoring config until /reload-mcp), and nothing explains concepts Claude already knows. Not the score-4 anchor since no section reads as trimmable padding.

5 / 5

Actionability

Guidance is copy-paste executable: the perl alarm probe, the CLI init/setup invocations with exact flags, sqlite3/discrawl id queries, the nohup daemon start, and a working config-merge snippet with backup. The one templated line (the commented ds.setdefault entry) is justified because entries depend on probe results.

5 / 5

Workflow Clarity

Steps 0-6 are explicitly sequenced with validation checkpoints and feedback loops: step 1a probes readability and stops on failure, step 3 proves each datasource syncs and removes failing ones, step 6 is a verification checklist, and Troubleshooting covers error recovery — matching the score-5 anchor including feedback loops.

5 / 5

Progressive Disclosure

scripts/autorag_inventory.py is a real, clearly signaled one-level-deep reference used at the right moment, and the body keeps overview-level content inline. However, the datasource table points to "entry shape in docs/datasource-skills.md", a file absent from the bundle — a clearly signaled but dangling reference, a minor organization gap that keeps it below the score-5 anchor.

4 / 5

Total

19

/

20

Passed

Description

92%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.

A strong description: it states the complete setup workflow in concrete verbs and gives explicit, specific trigger conditions drawn from the tool's own diagnostics. The only weakness is that trigger phrasing is machine-diagnostic rather than natural-user-language, costing it the top trigger-term score.

DimensionReasoningScore

Specificity

The description enumerates the full concrete workflow — "pick folders, write the config, auto-connect the Slack, Discord, Telegram, WhatsApp, KakaoTalk, and email archives already on this machine, build the first index, and prove a search works" — comprehensive coverage of the skill's actions, matching the score-5 anchor rather than the minor-gaps anchor at 4.

5 / 5

Completeness

Both what and when are explicit: a complete action list plus "Use when autorag.status or a search says index-not-ready, no folders are configured, a datasource is missing, search is keyword-only, or the autorag tools are absent" — concrete trigger phrases, matching the score-5 anchor.

5 / 5

Trigger Term Quality

Triggers are concrete and diagnostic ("index-not-ready", "no folders are configured", "search is keyword-only", "autorag tools are absent") and map to what a user would report, but they read as machine states rather than natural user synonyms (e.g., "set up AutoRAG", "autorag isn't working"), so coverage is good but not comprehensive.

4 / 5

Distinctiveness Conflict Risk

It occupies a clear niche (AutoRAG plugin setup) anchored by product-specific tool names and diagnostics ("autorag.status", "index-not-ready"), making conflict with other skills minimal.

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/setup/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.