CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

schedule

Keep the AutoRAG index fresh and watch for duplicate files on their own, using Hermes cron. Use when the user wants indexing to run automatically ("re-index every night", "only while charging", "only between 10pm and 7am") or wants to be told when duplicate documents appear ("check my Documents for duplicates every morning"), or wants to list or remove such schedules. Not for one-off searches (use the search skill), not for first-time setup (use the setup skill), and not for reacting to a file the moment it appears (use the automate skill).

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

92%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.

A tight, script-driven skill body: every command is copy-paste ready and verified against the actual bundle script, error handling and verification paths are explicit, and the read-only safety posture is stated up front. The only deduction is a pinned package version in the body, which is time-sensitive and will age.

DimensionReasoningScore

Conciseness

Lean and dense throughout — tables for job kinds and troubleshooting, flag semantics in terse bullets, no explanations of concepts Claude already knows — but the body pins a specific package version ("--autorag-cli npx -y -p @autorag/librarian@2.13.12 autorag"), which is time-sensitive information outside an old-patterns/deprecated section. Per the rubric's time-sensitivity guideline this is a minor conciseness penalty.

4 / 5

Actionability

Fully executable copy-paste commands for all four subcommands (index, duplicates, list, remove) with complete flag documentation, defaults, an explicit error gate ("If it exits non-zero, show the user its `error` and stop; do not retry with guessed flags"), and a fallback for a missing AutoRAG CLI. The referenced bundle script exists and implements exactly these commands and flags.

5 / 5

Workflow Clarity

Clear sequence ("Ask for what is missing, then run the helper. Do not guess a schedule the user did not give") with explicit validation (check the helper's JSON output and error exit), verification paths (`list`, `hermes cron status`), idempotency guidance (same `--id` updates rather than duplicates the job), and a symptom→check troubleshooting table as feedback loops. Operations are read-only by design ("It never moves, edits or deletes anything"), so the destructive/batch validation cap does not apply.

5 / 5

Progressive Disclosure

Well-organized sections (job-kind table, setup, check/list/remove, user-messaging, troubleshooting) with implementation detail correctly split into the single real bundle file, referenced one level deep via "${HERMES_SKILL_DIR}/scripts/autorag_schedule.py". Nothing that belongs in a separate file is inlined and navigation is easy.

5 / 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.

A strong description: it states concrete capabilities in third person, gives an explicit "Use when…" clause with quoted natural user phrases for both job kinds plus schedule management, and closes with explicit negative boundaries naming the sibling skills to use instead. The only soft spot is mildly figurative action phrasing ("keep the index fresh") rather than fully concrete verbs.

DimensionReasoningScore

Specificity

Lists several concrete actions — "Keep the AutoRAG index fresh", "watch for duplicate files", "list or remove such schedules" — but "keep the index fresh" is slightly figurative and the actions are not spelled out to the level of the anchor-5 example. It clearly exceeds the 1-2-action level of anchor 3.

4 / 5

Completeness

Explicitly answers both: what ("Keep the AutoRAG index fresh and watch for duplicate files on their own, using Hermes cron") and when ("Use when the user wants indexing to run automatically… or wants to be told when duplicate documents appear… or wants to list or remove such schedules") with concrete trigger phrases.

5 / 5

Trigger Term Quality

Quotes natural user utterances directly — "re-index every night", "only while charging", "only between 10pm and 7am", "check my Documents for duplicates every morning" — covering scheduling, power/window constraints, duplicate detection, and schedule management. No common phrasing variant is missing.

5 / 5

Distinctiveness Conflict Risk

Occupies a clear niche (scheduled no-agent cron jobs) and explicitly disambiguates from neighboring skills: "Not for one-off searches (use the search skill), not for first-time setup (use the setup skill), and not for reacting to a file the moment it appears (use the automate skill)". Minimal conflict risk.

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/schedule/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.