CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

writing-fragments

Writing, explore: mine raw fragments, no structure yet.

Quality

74%

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

Fix and improve this skill with Tessl

tessl review fix ./.agents/skills/writing-fragments/SKILL.md

The canonical home for this skill is writing-fragments in mattpocock/skills

SKILL.md
Quality
Evals
Security

Quality

Content

86%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 tight, concrete operating manual: it defines the skill-specific concept of a fragment with a valuable taxonomy, gives an exact file format, and specifies the runtime discipline (append-only, re-read before write, preserve user edits). Its only weaknesses are mild — a slightly indulgent example list and a workflow whose steps are scattered across two sections rather than presented as one ordered flow.

DimensionReasoningScore

Conciseness

The body is lean and assumes competence — no padding explaining what markdown or writing is — and the fragment taxonomy and file-format example are non-obvious, load-bearing content. Minor over-explanation remains: the 9-bullet fragment example list and the 'novelist's diary' aside could each be trimmed without losing guidance.

4 / 5

Actionability

For an instruction-only skill, the guidance is fully executable: concrete interview behavior ('grilling session... interviewing the user relentlessly'), a precise first-write format ('a single H1... no metadata, no TOC, no date'), an explicit save-path protocol, a copy-paste-ready file format with the exact separator ('\n---\n'), and first-class handling of concrete edit commands ('cut the last one', 'merge those two').

5 / 5

Workflow Clarity

The sequence is clear and unambiguous (grill for fragments → ask for the save path once → first write with H1 only → append silently thereafter), and it includes a genuine checkpoint: 'Before every write: re-read the file from disk... preserve their changes. Never overwrite; only append.' It falls short of 5 because the steps are distributed across the what-to-do and writing-rhythm sections rather than presented as one coherent ordered flow, leaving minor sequencing gaps.

4 / 5

Progressive Disclosure

This is a single-file skill with no bundle files (references/, scripts/, assets/ are absent), and none are needed: the body is short, all content is load-bearing on every turn, and it is cleanly organized (what-to-do vs supporting-info, with headed sections 'What is a fragment', 'File format', 'Writing rhythm'). Per the simple-skill scoring note, well-organized sections with no nested references score full marks.

5 / 5

Total

18

/

20

Passed

Description

50%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 terse and stylistically distinctive, but it describes only the 'what' in compressed, semi-private vocabulary. It lacks any explicit trigger guidance ('Use when...') and omits natural synonyms users would say when they need this skill, leaving it at the midpoint of the scale. (Note: with disable-model-invocation set, triggering is manual, which lowers the practical cost of the missing 'when' clause but does not change what the text states.)

Suggestions

Add an explicit trigger clause, e.g. 'Use when the user wants to explore or brainstorm raw material for a piece of writing — fragments, half-thoughts, quotes, vignettes — before any structure or outline exists.'

Include natural synonyms users would actually say — brainstorm, freewrite, 'I have scattered ideas/notes for an essay' — so the skill is recognizable from user phrasing.

Unpack 'mine raw fragments' slightly: state the concrete action (e.g. 'interview the user and capture fragments — sentences, claims, quotes, analogies — into a single markdown file').

DimensionReasoningScore

Specificity

The description names the domain ("Writing") and one concrete action ("mine raw fragments"), with "no structure yet" acting as a scope constraint rather than an action — matching the anchor for 'names domain and 1-2 concrete actions, but not comprehensive'. It is not a 4 because it does not list several specific actions.

3 / 5

Completeness

It has a clear 'what' (mine raw fragments, withhold structure) but no 'Use when...' clause or equivalent trigger guidance, which caps completeness at 3 per the judging guidelines. This is partially mitigated by disable-model-invocation (manual-only invocation), but the description text itself offers no explicit 'when'.

3 / 5

Trigger Term Quality

"Writing" and "fragments" are relevant keywords, but common natural phrasings users would actually say (brainstorm, freewrite, draft an essay, capture ideas) are missing. It sits above the 'one or two generic keywords' anchor because "fragments" is domain-specific rather than purely generic.

3 / 5

Distinctiveness Conflict Risk

"Writing" is very broad and overlaps with many conceivable writing-related skills, but "mine raw fragments, no structure yet" carves out a recognizable early-stage/explore niche. It could still overlap with brainstorming or outlining skills, matching the 'somewhat specific but could still overlap' anchor.

3 / 5

Total

12

/

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

tonykay/transcriber

Path

.agents/skills/writing-fragments/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.