CtrlK
BlogDocsLog inGet started
Tessl Logo

tpscloud/charmedimsure

Source-verified skills for the Charm terminal stack: Bubble Tea v2, Bubbles v2, Lip Gloss v2, Huh v2, Log v2, Glamour v2, plus the gum and vhs CLIs.

100

1.02x
Quality

100%

Does it follow best practices?

Impact

100%

1.02x

Average score across 9 eval scenarios

SecuritybySnyk

Passed

No findings from the security scan

Overview
Quality
Evals
Security
Files

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 body is a lean, high-density reference: executable examples for every subcommand, non-obvious behavior (stderr UI, abort exit statuses, pipe color stripping) stated as rules, a validated workflow with syntax and edge-case checks, and version-sensitive material isolated in an 'Older versions' section. Progressive disclosure is correctly implemented against the single one-level-deep reference file.

DimensionReasoningScore

Conciseness

Every section carries gum-specific facts Claude would not reliably know (exit status 130 on Ctrl+c, prompts drawing on stderr, ANSI stripping when piped, 2.x key bindings), with zero space spent on concepts Claude already knows. Time-sensitive version material is quarantined in an "Older versions" section and a baseline declaration, which the guidelines explicitly exempt from a conciseness penalty. The only trim candidates are one or two duplicated example lines between "Prompts" and the final recipe, which is composition value rather than padding — this fits the 5 anchor better than the 4 anchor ("minor instances of over-explanation").

5 / 5

Actionability

All code is copy-paste executable: `format=$(gum choose --header "Format" JSON YAML TOML) || exit 1`, `gum spin --spinner dot --show-output --title "Listing files" -- find . -type f`, `gum table --print --border rounded --file status.csv`, plus a complete runnable script under `set -euo pipefail`. Specific gotcha handling ("add `--show-error` to print it", "wrap it in `timeout(1)`") covers the common failure cases, matching the 5 anchor (fully executable, common cases covered).

5 / 5

Workflow Clarity

The 5-step workflow sequences version guarding (`gum version-check '>=2.0.0' || exit 1`), a command-selection decision map, capture/exit handling, syntax validation (`bash -n script.sh`), and edge-case testing ("Ctrl+c, Esc, a declined confirm, and an empty selection") — an explicit checklist. It includes a feedback loop: "If a prompt does not show or a value is wrong, check the rules in [How gum behaves] and the command's section in references/commands.md". Script writing is not a destructive or batch operation, so no cap applies; this matches the 5 anchor (explicit validation steps, error-recovery pointers, checklist).

5 / 5

Progressive Disclosure

Scored against the actual bundle: one reference file (references/commands.md, 207 lines) exists and contains sections for every subcommand named in the body; no scripts/ or assets/ are present or needed. It is referenced four times via clear markdown links ("Full flags and keys for every subcommand are in [references/commands.md]"), with the body enumerating exactly what lives there (style flags and env defaults, join layouts, format types, interactive table selection, pager search, log levels) — one level deep, well signaled, and easy to navigate ("If the request names a subcommand, go straight to it"). Exhaustive flag detail is delegated while examples and gotchas stay inline, matching the 5 anchor.

5 / 5

Total

20

/

20

Passed

Description

100%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.

An exemplary description: it states the complete command surface with what is covered (flags, exit codes, env defaults, examples), gives an explicit 'Use when' clause full of natural trigger phrases, and cleanly fences off the Go-library skills it could be confused with. Third person, dense, and free of fluff.

DimensionReasoningScore

Specificity

Enumerates the full concrete command surface — "choose, filter, confirm, input, write, file, spin, table, pager, style, join, format, and log, with their flags, exit codes, environment variable defaults, and Bash examples" — 13 named subcommands plus what coverage includes. This is comprehensive coverage of specific concrete actions, matching the 5 anchor; it goes beyond the 4 anchor ("several specific actions; minor gaps") because no meaningful capability of the tool is omitted. It is in third person ("Write interactive shell scripts"), so no voice penalty applies.

5 / 5

Completeness

Explicitly answers both: what — "Write interactive shell scripts with the gum CLI ... with their flags, exit codes, environment variable defaults, and Bash examples"; when — "Use when a Bash or shell script needs a prompt, menu, picker, fuzzy finder, yes or no question, text input, spinner, CSV table, pager, styled output ... or log lines". It also adds negative scope ("Not for Go programs (charmed-huh, charmed-bubbletea)"). This clearly matches the 5 anchor (both what and when with concrete trigger phrases), not the 4 anchor where the 'when' could be more specific.

5 / 5

Trigger Term Quality

Natural user phrasing is comprehensive: "prompt, menu, picker, fuzzy finder, yes or no question, text input, spinner, CSV table, pager, styled output with borders and terminal colors, rendered Markdown, or log lines", plus tool-name synonyms ("gum CLI (Charm, charmbracelet)") and shell terms ("Bash or shell script"). These are the words a user would actually say; the only conceivable additions (e.g. "checkbox", "multi-select") are covered by the listed synonyms. File extensions are not applicable to a CLI tool, so the anchor's extension criterion is satisfied in spirit.

5 / 5

Distinctiveness Conflict Risk

A clear niche — the gum CLI for shell-script prompting — with explicit disambiguation from adjacent skills: "without writing Go. Not for Go programs (charmed-huh, charmed-bubbletea)". Naming the specific tool and its vendor makes wrong-skill triggering very unlikely, matching the 5 anchor ("Clear niche with distinct triggers; minimal conflict risk").

5 / 5

Total

20

/

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.

Reviewed

Table of Contents