Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/melodic-software/claude-code-plugins/reducenpx skills add melodic-software/claude-code-plugins --skill reducegit clone --depth 1 https://github.com/melodic-software/claude-code-pluginsWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00192 | $0.02728 |
| Opus 5 | $0.00096 | $0.01364 |
| Sonnet 5 | $0.00038 | $0.00546 |
| Haiku 4.5 | $0.00019 | $0.00273 |
Grade A, and why
reduce scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 190 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Pre-computed context
Current branch: !git branch --show-current 2>/dev/null || echo "unknown"
Recent commits: !git log --oneline -10 2>/dev/null || echo "no commits"
Working tree status (empty = clean): !{ git status --porcelain 2>/dev/null || echo "(git status unavailable)"; } | head -10
Variables
Arguments: $ARGUMENTS
Purpose
Answer one question, repeatedly: which dependency in this repository transmits the most unnecessary change, and what is the smallest mechanism that stops the transmission? Each invocation is one pass. Scan, verify, reduce what is safely reducible within a scope budget, surface what is not, and record everything in a durable ledger so the next invocation resumes instead of restarting. Coupling goes down monotonically across runs; no single run tries to finish the job.
Two lanes, split by the nature of the finding, never by convenience:
- Apply lane. Mechanical, contained, behavior-preserving reductions (weaken a connascence form, de-duplicate a stated fact into a pointer, inject a hard-wired volatile collaborator, name a magic value). Applied this run, verified, shipped as one structure-only change set.
- Route lane. Cross-file remediation and architectural judgment (move a boundary, introduce a seam, split a shared database, merge two repos' duplicated logic). Coupling findings of this shape exist to inform a human: they are surfaced, ranked, and routed, never auto-applied.
This is not a diff reviewer, not a designer of one chosen boundary, and not a general tidy pass. See "What this skill does NOT do".
Actions
| Argument | Action |
|---|---|
| (empty) | Full pass over an inferred scope: scan → verify → apply lane → route lane → ledger |
<scope> |
Same pass narrowed to a path, module, or altitude keyword (docs, code, app, repo) |
dry-run [<scope>] |
Scan, verify, rank, and ledger only. Do NOT edit, branch, push, or file tracker items |
status |
Read the ledger and report open candidates, applied reductions, and what the next run should take |
help |
Print this table with one worked example per action |
What ships with it
4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 190 lines · 192 tokens per session scan A 7dc908932e1b
reduce is a skill published in the GitHub repository melodic-software/claude-code-plugins (12 stars, last pushed 2d ago), licensed MIT. It adds 192 tokens to every session and 2,728 once invoked, about $0.0010 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
deep-research
Conducts iterative deep research on any topic using web search, progressive exploration, and structured synthesis. Use when asked for comprehensive research, deep investigation, thorough analysis, or multi-source exploration of any topic. Triggers: research, investigate, deep dive, comprehensive analysis, explore…
error-ux
Principles and patterns for writing error messages that help users recover. Use when auditing, writing, or improving error messages in code. Triggers: error messages, user experience, error handling, exception messages, validation errors.
adversarial-patterns
Library of realistic adversarial attack vectors and anti-patterns to avoid. Contains examples of valid attacks and subtle gaming patterns to reject.
documentation-testing
Provides heuristics for identifying incomplete or broken documentation. Use when validating README setup instructions, testing onboarding flows, or auditing documentation quality. Triggers: docs, readme, onboarding, setup validation, documentation audit.
adversarial-analysis
Analyze code to identify explicit contracts, implicit usage patterns, and realistic boundary conditions. Contains concrete formulas for calculating input realism limits. Use before generating adversarial tests.
propagate-then-search
For constraint problems: eliminate impossibilities before guessing, reduce search space through inference, fail fast on contradictions.