CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

batch-api

Prepare a many-file, single-turn transform (translate, rewrite, extract) as a plan and submit it to the asynchronous, half-price DashScope Batch API; results are delivered as new files hours later. Invoke explicitly with /batch-api.

Quality

88%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

88%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, a complete plan template, and a validation-gated workflow (check → preview → submit with digest → background collect) with explicit error-recovery loops. The only refinements are consolidating the repeatedly stated never-do-it-yourself rule and potentially moving the plan schema detail to a reference file.

DimensionReasoningScore

Conciseness

The body is dense and operation-specific — CLI subcommands, refusal rules, the plan-file schema — with essentially no explanation of concepts Claude already knows. Not 5: the 'never do the work yourself / never bill without a preview' rule is restated across the preamble, §4, and §5 ('Never redo a failed item yourself in this session') and could be consolidated; not 3 because the trimming opportunity is minor.

4 / 5

Actionability

Every step ships exact, copy-paste-ready commands — '"${QWEN_CODE_CLI:-qwen}" batch check', 'batch run .qwen/batch/plans/<slug>.json --dry-run', 'run ... --expect <digest>', 'collect <task-id> --wait' — plus a complete JSON plan template with field-level rules. Not 4: no gaps; the placeholders (<slug>, <task-id>) are the only variables and are defined inline.

5 / 5

Workflow Clarity

§0–§5 give a clear sequence with explicit validation checkpoints for a billed batch operation: an unbilled readiness check, an unbilled dry-run preview, digest pinning via '--expect', and two feedback loops (preview mismatch → re-preview and reshow; plan-field error → fix once and preview again). Not 4: validation is not merely present but woven into every spending decision, matching the validate→fix→retry anchor.

5 / 5

Progressive Disclosure

No bundle files exist (references/, scripts/, assets/ are all absent) and the single SKILL.md is well organized into numbered sections with no nested or dangling references, so navigation is trivial. Not 5: at roughly 190 lines the plan-file schema and its field rules are all inline where a reference file could hold the detail; not 3 because structure is clear and nothing is buried.

4 / 5

Total

18

/

20

Passed

Description

83%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: concrete, third-person, and comprehensive about what the skill does, with an explicit invocation cue and a distinct DashScope Batch niche. Its only weaknesses are trigger-term synonym coverage (no 'bulk'/'convert' variants) and a 'when' clause that says how to invoke rather than when the skill is the right choice.

Suggestions

Add natural trigger synonyms such as 'bulk', 'in bulk', or 'convert' alongside 'translate, rewrite, extract' so users phrasing the need conversationally still match the skill.

Extend the closing clause to state when the skill is the right choice, e.g. 'Invoke explicitly with /batch-api when many files need the same transform and the result can wait hours for a half-price run.'

DimensionReasoningScore

Specificity

The description lists multiple concrete actions — 'Prepare a many-file, single-turn transform (translate, rewrite, extract) as a plan', 'submit it to the asynchronous, half-price DashScope Batch API', 'results are delivered as new files hours later' — with comprehensive coverage of the skill's capability. Not 4: there are no meaningful coverage gaps for this niche; not below 5 because every listed action is concrete and specific.

5 / 5

Completeness

Has a clear 'what' (plan-and-submit workflow for the Batch API with file delivery) and explicit trigger guidance ('Invoke explicitly with /batch-api'), which satisfies the equivalent-trigger clause so the cap at 3 does not apply. Not 5: the 'when' states how to invoke rather than when this skill is the right choice (e.g., 'use when you need to transform many files and can wait hours for half price').

4 / 5

Trigger Term Quality

Good natural keyword coverage — 'translate, rewrite, extract', 'asynchronous', 'half-price', 'batch', 'plan' — plus the explicit '/batch-api' invocation cue. Not 5: common user phrasings like 'bulk', 'in bulk', or 'convert' are absent; clearly above 3 because the core terms users would say when needing this skill are present.

4 / 5

Distinctiveness Conflict Risk

'DashScope Batch API' and 'Invoke explicitly with /batch-api' carve out a clear niche with distinct triggers and minimal overlap risk with general transform or file skills. Not 4: the provider-specific naming plus explicit-only invocation leaves no meaningful overlap with related skills.

5 / 5

Total

18

/

20

Passed

Validation

93%

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

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

15

/

16

Passed

Repository

QwenLM/qwen-code

Path

packages/core/src/skills/bundled/batch-api/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.