CtrlK
BlogDocsLog inGet started
Tessl Logo

hunk-extensions

Maps the `hunkdiff/extension` authoring surface for Hunk, the terminal diff viewer — hiding or reordering reviewed files, docked panes, alternate file views, commands and key bindings, dialogs, workspace writes, themes, syntax languages, VCS backends, lifecycle events. Use when writing, debugging, or installing a Hunk extension, or when a request asks Hunk itself to behave differently. Not for reviewing a diff in a live session — that is hunk-review.

67

Quality

81%

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

75%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 well-organized, high-signal map of the extension authoring surface with executable examples, a ranked verification workflow, and clear navigation to source-of-truth files. Its main gaps are the absence of bundle files to offload reference-grade detail and slightly less explicit validation loops for destructive paths.

Suggestions

Add a references/ file (e.g. API.md) for the 'What handlers receive' ctx field inventory and 'Rules that bite' list so SKILL.md stays a lean overview, then link to it from the touchpoint table.

Wrap the `ctx.workspace` write guidance in an explicit validate-checkpoint (check `canWriteDocument` → write → confirm `{ ok }` → else handle `reason`) so destructive operations get the missing feedback loop.

Add one inline copy-paste snippet per common touchpoint (e.g. a minimal `registerPane` example) alongside the table so the most frequent cases are fully executable without opening the external docs.

DimensionReasoningScore

Conciseness

The body is dense and almost entirely non-obvious domain knowledge (id rules, trust gating, lifetime semantics, the second-React bug), with little padding of concepts Claude already knows. It earns a 4 rather than 5 because sections like 'Where extensions live' and the examples catalog are information-rich but could be tightened, and a few sentences restate constraints already implied by the table.

4 / 5

Actionability

Provides executable code (the hello factory), concrete commands (`bun run typecheck`, `hunk --extension ./my-ext.ts my-command sync --help`), and a precise API lookup table mapping intent to call. Not a 5 because most guidance is a map to external sources rather than inline copy-paste snippets for each common case.

4 / 5

Workflow Clarity

The 'Verifying' section is an ordered, cost-ranked checklist with explicit checkpoints, and 'If it does not load' gives a feedback loop for triage. It stops at 4 because the build path lacks an explicit validate-then-retry loop for the typecheck/JSX-config steps, and the destructive-workspace-write rules are stated but not wrapped in a verify checkpoint.

4 / 5

Progressive Disclosure

Structure is good: a touchpoint table plus a 'Sources of truth — read before writing' table that signals one-level-deep references (docs/extensions.md, types.ts, examples/) for detail. No bundle files exist in references/scripts/assets to offload the API reference, so some reference-grade content is inlined; it stays at 4 rather than 5 because the inlined 'What handlers receive' and 'Rules that bite' blocks read like reference material that a separate file could hold.

4 / 5

Total

16

/

20

Passed

Description

87%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 specific, well-triggered, and clearly distinct from the sibling hunk-review skill via an explicit exclusion clause. Trigger-term coverage is strong but could add a few more natural synonyms or a concrete file-type cue.

DimensionReasoningScore

Specificity

Lists several concrete actions — 'hiding or reordering reviewed files, docked panes, alternate file views, commands and key bindings, dialogs, workspace writes, themes, syntax languages, VCS backends, lifecycle events' — giving broad coverage of the authoring surface. Falls just below a 5 because the enumeration emphasizes surfaces rather than discrete end-user actions (e.g. 'add a navigation pane' is implied, not stated).

4 / 5

Completeness

Explicitly answers both 'what' (maps the authoring surface, enumerating the touchpoints) and 'when' ('Use when writing, debugging, or installing a Hunk extension, or when a request asks Hunk itself to behave differently'), plus a concrete exclusion ('Not for reviewing a diff in a live session — that is hunk-review').

5 / 5

Trigger Term Quality

Includes natural phrases a user would say — 'writing, debugging, or installing a Hunk extension' and 'asks Hunk itself to behave differently' — plus the domain terms. Not a 5 because it lacks common synonym variations and an explicit file-type/extension trigger like '.ts extension'.

4 / 5

Distinctiveness Conflict Risk

Occupies a clear niche (the `hunkdiff/extension` authoring surface) and disambiguates from the sibling skill via an explicit negative trigger ('Not for reviewing a diff in a live session — that is hunk-review'), minimizing conflict risk.

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.

Validation15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

referenced_paths_exist

Referenced path issues: 1 missing

Warning

Total

15

/

16

Passed

Repository
modem-dev/hunk
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.