CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

cordis-plugin-development

Use when designing, reviewing, adding, enabling, disabling, installing, configuring, or debugging a plugin, bundle, feature, page, panel, tool, or MCP connection in the current Harness profile, including a shipped plugin that is disabled by default, and for any visual object, decoration, or widget request that names no other destination, which means an installed UI plugin rendered in the Harness Web UI.

Quality

87%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Medium

Suggest reviewing before use

SKILL.md
Quality
Evals
Security

Quality

Content

100%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 content is an exemplary instruction-only skill: token-efficient, dense with exact tool names, parameters, and paths, and a numbered delivery workflow with real validation checkpoints and a repair loop. Reference material is appropriately externalized into a complete, one-level-deep file table that matches the actual bundle.

DimensionReasoningScore

Conciseness

The body is lean and imperative with no padding or explanation of concepts Claude already knows; dense operational sentences like 'PowerShell, which the Windows preset uses, reads them as $env:DSH_PROFILE' carry only new information. Every token earns its place across the 44-line body.

5 / 5

Actionability

Guidance is fully executable for an instruction-only skill: exact tool calls with parameters ('plugin_manager with action: install_bundle and the absolute package directory as target', 'cordis_inspect_list, then targeted cordis_inspect_query'), exact paths ('packages/bundle/*/cordis.patch.yml'), and a runnable command ('dsh --profile "$DSH_PROFILE" --dump-config'). Not 4: concrete specifics cover the common cases, with template files supplying the copy-paste code.

5 / 5

Workflow Clarity

The 'Deliver a working plugin first' section is a clear 5-step sequence with explicit validation checkpoints ('application and warnings decide whether the change is live', 'Confirm new rows with approval-free cordis_inspect_query'), a fix-defects feedback loop (step 5), a display checklist, user-state restoration, and honest verification limits. Not 4: checkpoints and error recovery are explicit rather than minor-gap.

5 / 5

Progressive Disclosure

A task-to-file 'Read next' table declared as the complete list gives clear navigation, all six references/*.md files exist in the bundle with substantive content, and references stay one level deep (cross-pointers to templates and sibling references, no nested chains). Not 4: the split between overview, knowledge sources, and per-topic references is exemplary.

5 / 5

Total

20

/

20

Passed

Description

70%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 a strong, highly explicit trigger statement with concrete domain vocabulary and excellent distinctiveness, but it is one-sided: it answers 'when to use' exhaustively while never stating what the skill actually does. Adding a brief capability clause before the 'Use when' clause would make it complete.

Suggestions

Add a leading 'what' clause stating the skill's capabilities, e.g. 'Authors and installs Harness plugin bundles via plugin_manager install_bundle... Use when...' — currently the skill's function must be inferred from trigger verbs.

Include natural trigger variations users are likely to say, such as 'uninstalling' or 'removing' a plugin and possibly 'extension' or 'add-on'.

Trim the trailing definitional clause ('which means an installed UI plugin rendered in the Harness Web UI') into the main trigger list to reduce length without losing the fallback trigger for unnamed visual requests.

DimensionReasoningScore

Specificity

Lists many concrete actions ('designing, reviewing, adding, enabling, disabling, installing, configuring, or debugging') and concrete objects ('plugin, bundle, feature, page, panel, tool, or MCP connection', 'the Harness Web UI'). Not 5: it never explicitly states what the skill itself does (authoring bundles, installing them via plugin_manager); not 3: coverage is broad and concrete rather than 1-2 actions.

4 / 5

Completeness

The 'when' is explicit and comprehensive ('Use when designing, reviewing, ... in the current Harness profile'), but the 'what does this skill do' half is never stated — it is only implied by the trigger verbs. This falls between anchor 2 (only 'when' without 'what') and a fully answered pair, with the 'when' side far stronger than anchor 2's example.

3 / 5

Trigger Term Quality

Good coverage of natural terms a user would say — 'plugin', 'install', 'enable', 'disable', 'configure', 'debug', 'widget', 'decoration', 'page', 'panel', 'MCP connection'. Not 5: a few natural variations are missing (e.g., 'uninstall'/'remove' a plugin, 'extension'); not 3: synonyms across object types are well covered.

4 / 5

Distinctiveness Conflict Risk

Clear niche with distinct triggers: everything is scoped to 'the current Harness profile' and 'the Harness Web UI', and the plugin/bundle/MCP-connection vocabulary is unlike generic skill territory. Minimal conflict risk with other skills.

5 / 5

Total

16

/

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

Hmbown/Codewhale

Path

crates/skills/assets/dsh-creator/skills/cordis-plugin-development/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.