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 skills add iuliandita/skills --skill code-simplificationgit clone --depth 1 https://github.com/iuliandita/skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/iuliandita/skills/code-simplification)<a href="https://agentmods.dev/skills/iuliandita/skills/code-simplification"><img src="https://agentmods.dev/badge/skills/iuliandita/skills/code-simplification/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/iuliandita/skills/code-simplification"><img src="https://agentmods.dev/badge/skills/iuliandita/skills/code-simplification.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What 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 5.1 | $0.00030 | $0.01578 |
| Opus 5.5 | $0.00012 | $0.00631 |
| Sonnet 5.5 | $0.00006 | $0.00316 |
| Haiku 4.5 | $0.00003 | $0.00158 |
Grade A, and why
code-simplification 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 5d 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 — 84 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Code Simplification
Review code for unnecessary complexity, ungrounded implementation choices, and safe reductions. Preserve behavior and report evidence in maintainability or reduction mode.
- Maintainability: over-abstraction, generic naming, noisy comments, invented APIs or configuration, stale idioms, dependency creep, and weak tests.
- Safe reduction: dead or superseded code, orphan files, exact clones, redundant wrappers, inert defensive scaffolding, and comment walls.
Smaller code is not automatically better. A finding needs a concrete maintenance or correctness cost, an invariant, and evidence that the proposed change preserves contracts.
When to use
- Review code or a PR for unnecessary complexity, AI-shaped code, grounded API use, and useful tests.
- Find behavior-preserving reductions before a cleanup change.
- Check a removal or refactor diff for lost behavior, fallbacks, or reachable entry points.
When NOT to use
- Find logic bugs, races, or regressions: use code-review.
- Find vulnerabilities or secrets: use security-audit.
- Rewrite prose for voice: use anti-ai-prose.
- Run a repository-wide multi-lens audit: use repo-audit.
- Implement a cleanup: use the relevant language or framework skill.
Workflow
- Scope: prefer the named path or diff; otherwise inspect uncommitted changes, then the current branch against its confirmed integration base. Record generated, vendored, and unreviewed areas.
- Establish context: read instructions, manifests, runtime versions, public exports, registries, route/CLI/plugin discovery, and existing validation. For maintainability mode, load
references/maintainability-patterns.mdand each relevant language or infrastructure reference; for reduction mode, loadreferences/patterns.mdandreferences/reduction-patterns.md. Loadreferences/research-sources.mdonly when citations help. - Make a maintainability pass: inspect scripts and configuration before running any baseline tooling. Run only configured, non-mutating check commands; never use
--fix,--write,apply, or an unknown script. Then inspect comments, naming, types, wrappers, error handling, dependencies, API/config claims, and tests (including test bloat, see Rules). Verify suspicious APIs with local types, schemas, lockfiles,--help, or primary documentation before calling them invented. - Make a safe-reduction pass: search for dead and superseded code, leftovers, clones, one-literal-per-function copies, redundant parsers, thin wrappers, and inert catches. Read each candidate, a caller, and relevant tests.
- Prove or defer: search with
rg -n(usegrep -rnwhencommand -v rgfails). A deletion needs no-reference proof that rules out public APIs, reflection, dynamic dispatch, DI, serialization, route/CLI/plugin registration, conditional builds, and test discovery. Preserve boundary guards, recoveries, retries, cleanup, observability, compatibility, and intentional provider divergence. - Classify: use
Do nowonly for mechanically proven, behavior-preserving reductions;Do with testswhen focused validation is missing;Deferfor broad, hot-path, or uncertain work;Leave alonewhen the current shape carries a contract. Also classify maintainability findings as Noise, Lies, or Soul and as Fix, Consider, or Fine (defined inreferences/maintainability-patterns.md); project conventions and framework idioms override generic style rules. - Report: write an audit deliverable using
references/report-templates.md. Group repeated examples into one finding and include location, evidence, invariant, proposed shape, tradeoff, and validation needed.
What ships with it
14 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.
- references/agent-hygiene.md 1.1 KB
- references/docker.md 7.1 KB
- references/iac.md 12 KB
- references/language-patterns.md 2.3 KB
- references/maintainability-patterns.md 3.3 KB
- references/output-contract.md 2.8 KB
- references/patterns.md 6.5 KB
- references/python.md 6.4 KB
- references/reduction-patterns.md 2.1 KB
- references/report-templates.md 4.5 KB
- references/research-sources.md 3.5 KB
- references/rust.md 5.4 KB
- references/shell.md 5.1 KB
- references/typescript.md 5.7 KB
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.
- 5d ago Changed · +1 lines 24e29c3cf8c5
- 14d ago Changed · -1 lines c06470658f8c
- 23d ago Changed · +1 lines 5c20d5dce4f2
- 27d ago First seen · 83 lines · 30 tokens per session scan A 648831b264e2
code-simplification is a skill published in the GitHub repository iuliandita/skills (7 stars, last pushed 3d ago), licensed MIT. It adds 30 tokens to every session and 1,578 once invoked, about $0.0001 per session on Opus 5.5. 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 2026-09-21.
Other skills, from other repositories
code-reviewer
Code review knowledge base: quality, security (OWASP Top 10), error-handling, performance, and test-coverage checklists with severity-ranked output format. Use when reviewing code changes, PRs, or before commits. Loaded automatically by the code-reviewer agent.
factory-ci
CI and pull-request review conventions. One canonical .github/workflows/ci.yml is the merge gate (typecheck, lint, test, build, claude-review); branch protection's required-checks list matches the job list 1:1; anthropics/claude-code-action@v1 reviews every PR against the factory-pitfalls.md checklist as a required…
factory-code-reviewer
Use to review a PR, diff, or recently-written code against the factory's conventions. Read-only — outputs structured review, not diffs. Carries the full factory-pitfalls digest as a PR checklist plus the conventions from every other factory-.md skill. Flags anti-patterns, missing conventions, security risks, and…
judge
Obtain an independent review of the approach and trade-offs in completed work.
audit
Review code through focused dimensions, with one independent agent per dimension.
code-review
Reviews code for design, correctness, performance, and test quality. Use when the user wants a code review of files, a diff or PR, or the whole project before shipping or merging.