dedup-skills

Solid

Use when asked to deduplicate skills or prompt directories, find repeated rules, or check a skill tree for contradictions. Produces a read-only ledger classifying every repetition cluster and conflict candidate with evidence and zero unclassified cells.

AI & Automation 38 stars 0 forks Updated yesterday Apache-2.0

Install

View on GitHub

Quality Score: 82/100

Stars 20%
53
Recency 20%
100
Frontmatter 20%
70
Documentation 15%
100
Issue Health 10%
50
License 10%
100
Description 5%
100

Skill Content

# Dedup skills ## Contract | Field | Bound contract | |---|---| | Trigger | User asks to deduplicate skills/prompts, find repeated rules, or check a skill tree for contradictions. | | Authority | Read-only: scans a markdown skill/prompt tree and emits a chat ledger. No file, VCS, credential, paid, published, deployed, or remote mutation. | | Side effect | Produces a repetition/conflict ledger in chat; leaves the scanned tree untouched. | | Done | Every repetition cluster and conflict candidate is classified repeat/conflict/intentional-keep/not-a-finding, with totals, evidence, and zero unclassified cells. | ## Inputs - Target tree path (default `skills/`). A prompt-directory tree may be supplied instead. - The tree's LICENSES/NOTICE/attribution file, if present, is read in full before scanning. ## Procedure 1. **Vendored set.** Read the tree's LICENSES/NOTICE/attribution file in full before any scan. Enumerate every vendored or license-covered path. Report findings inside vendored paths with a vendored annotation and never propose them for deletion. Exclude the attribution file itself from the scan because it is a registry, not prompt content. Done when: every vendored or license-covered path is enumerated and the attribution file is excluded from the scan. 2. **Cluster pass.** Run a mechanical script over all markdown in the tree. Strip YAML frontmatter first. Frontmatter is load-bearing routing data, never a dedup target, and must not enter the scan. Segment the text ...

Details

Author
OutlineDriven
Repository
OutlineDriven/odin-claude-plugin
Created
10 months ago
Last Updated
yesterday
Language
Python
License
Apache-2.0

Similar Skills

Semantically similar based on skill content — not just same category

AI & Automation Listed

dedupe-skills

Overlapping-skill auditor. Sweeps every installed skill surface (plugins, personal and project skills), clusters skills that do the same job, collide on triggers, contradict each other when both active, or burn always-on input tokens redundantly — then recommends which to keep and which to remove, and performs only the removals the user explicitly confirms (preferring reversible disable over uninstall). Trigger on /eff:dedupe-skills or $eff:dedupe-skills, and on natural-language asks like: 'find duplicate skills', 'which of my skills overlap', 'my skills conflict with each other', 'clean up redundant skills', '겹치는 스킬 찾아줘', '중복 스킬 정리해줘', '스킬끼리 충돌하는 것 같아', '비슷한 스킬 추려줘'. Keep this distinct from eff:update-skills (version bumps only) and eff:prune-skills (removal by low USAGE) — dedupe-skills removes by functional OVERLAP.

0 Updated 1 months ago
im-ian
AI & Automation Listed

skill-cleaner

Use when a skills registry feels messy or untrustworthy: a skill that seems to work some days and not others, a skill you wrote that never gets picked, duplicate or overlapping skills competing for the same task, skills scattered across projects and home directories, dead symlinks, or before publishing a skills repo. Also use when merging redundant skills down to one survivor, or when auditing SKILL.md frontmatter for spec compliance. Appends new failure modes to its own pattern list after each run.

0 Updated 3 days ago
heyramzi
AI & Automation Listed

skill-curator

Audit a skill library for duplicated guidance, contradictions, bloat, dead references, and defects. Use when cleaning, consolidating, or de-duplicating a skills tree.

2 Updated yesterday
wilbeibi