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 eristoddle/agent-skills --skill skill-hardenergit clone --depth 1 https://github.com/eristoddle/agent-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/eristoddle/agent-skills/skill-hardener)<a href="https://agentmods.dev/skills/eristoddle/agent-skills/skill-hardener"><img src="https://agentmods.dev/badge/skills/eristoddle/agent-skills/skill-hardener/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/eristoddle/agent-skills/skill-hardener"><img src="https://agentmods.dev/badge/skills/eristoddle/agent-skills/skill-hardener.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 | $0.00098 | $0.01924 |
| Opus 5 | $0.00049 | $0.00962 |
| Sonnet 5 | $0.00020 | $0.00385 |
| Haiku 4.5 | $0.00010 | $0.00192 |
Grade A, and why
skill-hardener 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 10d 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 — 140 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Skill Hardener
Turn frustrating sessions into permanent fixes. This skill reads your own Claude Code transcripts, finds the patterns that keep biting you, and proposes concrete guardrails — each paired with a regression test that proves the fix and catches the regression.
It is the diagnostic complement to skill-creator (which authors and evals skills). skill-hardener decides what to harden and why; it hands the actual edit + eval to skill-creator's machinery when that's the right tool.
The prime directive: this skill edits other skills, so it is dangerous
A tool that rewrites your skills can quietly degrade them. Every safeguard below is load-bearing — do not skip them to move faster:
- Never auto-apply. Always show a diff and get explicit approval before writing to any skill or CLAUDE.md. No exceptions.
- Recurrence gate. Only propose a guardrail for a pattern seen ≥3 times across ≥2 distinct sessions (default; user may lower). One bad session is an anecdote, not a pattern. Hardening against an anecdote is how skills rot into brittle patch-piles.
- Prefer general over specific. A guardrail should express a principle, not patch one transcript. If you can only describe the fix by referencing one session, it's not ready.
- Cap the blast radius. Max 3 guardrail edits per run. More than that means the diagnosis is too shallow — go narrower.
- Git is the undo button. Refuse to apply edits in a dirty/untracked skill dir without warning; every change must be a reviewable, revertible diff. Note the rollback command in the changelog.
- Don't harden what you can't attribute. If you can't tie a failure to a specific skill or to base/CLAUDE.md behavior, report it — don't guess at a target.
Workflow
Phase 1 — MINE & SHOW
Always lead with the plain-language, skill-first report. This is what the user reads: which skills gave them the most grief, ranked, with bars. Do NOT open with raw signature counts.
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.
- 10d ago First seen · 140 lines · 98 tokens per session scan A bc049aeca7d1
skill-hardener is a skill published in the GitHub repository eristoddle/agent-skills (2 stars, last pushed 15d ago), licensed MIT. It adds 98 tokens to every session and 1,924 once invoked, about $0.0005 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 2026-08-31.
Other skills, from other repositories
engram-testing-coverage
TDD and coverage standards for Engram. Trigger: When implementing behavior changes in any package.
iterative-development
TDD iteration loops using Claude Code Stop hooks - runs tests after each response, feeds failures back automatically.
python
Python development with ruff, mypy, pytest - TDD and type safety.
nw-fp-clojure
Clojure language-specific patterns, data-first modeling, REPL-driven development, and spec.
strict-tdd
Strict RED->GREEN->REFACTOR test-driven development with enforcement. Never write production code before a failing test. Atomic commits per TDD cycle.
mobiai-mobile-tdd
You MUST use this before writing any implementation code for a mobile feature, bug fix, refactor, or behavior change. Tests come before implementation — no exceptions.