CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

citation-fixer

Audit and fix citation formatting across brain pages. Ensures every fact has an inline [Source: ...] citation matching the standard format. Extended in v0.25.1: scans for broken tweet/post references that lack actual URLs and resolves them via the host's X / Twitter API integration.

Quality

71%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./skills/citation-fixer/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

67%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 a well-structured, highly actionable instruction document with concrete query patterns, executable batch tooling, explicit verification steps, and thoughtful failure recovery. Its weaknesses are changelog-style version repetition, a contentless duplicate "Output Format" section, and an implicit (rather than explicit) fallback for failed tweet searches.

Suggestions

Collapse the repeated "(v0.25.1)" markers into a single note; describe current behavior plainly and leave version history out of the body.

Delete or merge the trailing duplicate "Output Format" stub section — a section whose only content says it exists for a conformance test wastes tokens and confuses navigation.

Add an explicit fallback for tweet searches that return zero or multiple candidates (e.g. flag the reference unresolvable in the report, or narrow the query with the date and re-search).

Fix the batch grep pattern so "RT" is matched as a word boundary (e.g. grep -Ewi "retweet|RT") to avoid false positives from substrings like "part" or "start".

DimensionReasoningScore

Conciseness

The body is mostly operational, but padding is noticeable: the "(v0.25.1)" changelog marker is repeated six times, the tweet pipeline is labeled as a "v0.25.1 extension", and a second "Output Format" section at the end exists only "for the conformance test" with no content. This fits the anchor "mostly efficient but includes some unnecessary explanation or could be tightened"; it is not score 4 because the version narration and the dead duplicate section are clearly trimmable.

3 / 5

Actionability

Concrete guidance throughout: exact X search query patterns in code blocks, a copy-paste bash candidate-finder, before/after citation patch examples, an exact report template, and specific error-code handling (revision_conflict, page_not_found). It falls short of score 5 because the API surface is described only loosely ("whatever X integration the host has configured") and the batch grep token "RT" matches unintended substrings (e.g. "part", "start") under case-insensitive matching.

4 / 5

Workflow Clarity

A clear five-phase sequence plus a per-reference pipeline with explicit verification ("Confirm the text matches the quoted fragment"), batch-commit checkpoints every 10-20 pages, sample-check-before-bulk, and revision-conflict recovery steps. It is not score 5 because there is no explicit fallback when a tweet search returns zero or multiple candidates — that handling is only implied by the "remaining gaps" report field.

4 / 5

Progressive Disclosure

The body is well sectioned and points to one-level-deep, clearly signaled references at point of use (conventions/quality.md, _output-rules.md, conventions/agent-operator-protocol.md, conventions/test-before-bulk.md); no bundle files exist alongside this SKILL.md. Minor gaps keep it below 5: the tweet resolution pipeline detail could live in a reference file, and the duplicate "Output Format" header at the end hurts navigation.

4 / 5

Total

15

/

20

Passed

Description

63%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 and covers its capabilities comprehensively in third person, but it omits any "when to use" guidance within the description text itself, leaving natural trigger phrasings confined to the frontmatter triggers block. Adding an explicit use-when clause with the natural user phrasings would raise both completeness and trigger quality.

Suggestions

Append an explicit trigger clause to the description, e.g. "Use when the user asks to fix, check, or audit citations, or when brain pages contain tweet/post references without links."

Fold the natural user phrasings from the triggers block ("citation audit", "check citations") into the description body so the description alone carries trigger terms.

Replace changelog-style version narration ("Extended in v0.25.1:...") with a plain capability statement; version history belongs elsewhere.

DimensionReasoningScore

Specificity

"Audit and fix citation formatting across brain pages", "Ensures every fact has an inline [Source: ...] citation matching the standard format", and "scans for broken tweet/post references that lack actual URLs and resolves them via the host's X / Twitter API integration" list multiple specific concrete actions with comprehensive coverage of the skill's domain. It exceeds the score-4 anchor (several specific actions, minor gaps) because no meaningful capability is left out.

5 / 5

Completeness

The "what" is clear (audit/fix citation formatting, resolve broken tweet references), but the description contains no "Use when..." clause or equivalent explicit trigger guidance; the guideline caps completeness at 3 in that case. The separate frontmatter triggers list mitigates this in the full frontmatter, but the description field on its own fits the anchor "clear 'what' but 'when' is missing"; it is not score 2 because the what is specific and well stated.

3 / 5

Trigger Term Quality

The description contains some relevant natural keywords ("citation formatting", "broken tweet/post references", "X / Twitter") but misses common variations users would say, such as "check citations", "sources", or "broken links" — those phrases appear only in the separate triggers block, not in the description text itself. It fits the anchor "some relevant keywords but missing common variations or synonyms" rather than score 4, whose keyword coverage is explicitly broad.

3 / 5

Distinctiveness Conflict Risk

"Citation formatting across brain pages" with "[Source: ...]" conventions and "the host's X / Twitter API integration" defines a mostly distinct niche with minimal conflict risk. It is not score 5 because "citation" and "link fixing" terminology could still overlap with generic citation/bibliography or link-repair skills, matching the anchor "mostly distinct; minor overlap risk with closely related skills".

4 / 5

Total

15

/

20

Passed

Validation

87%

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

Validation — 14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

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

Warning

relative_links

Relative link issues: 3 suspicious

Warning

Total

14

/

16

Passed

Repository

garrytan/gbrain

Path

skills/citation-fixer/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.