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 MatrixFounder/Agentic-development --skill skill-creatorgit clone --depth 1 https://github.com/MatrixFounder/Agentic-developmentWrote 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/matrixfounder/agentic-development/skill-creator)<a href="https://agentmods.dev/skills/matrixfounder/agentic-development/skill-creator"><img src="https://agentmods.dev/badge/skills/matrixfounder/agentic-development/skill-creator/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/matrixfounder/agentic-development/skill-creator"><img src="https://agentmods.dev/badge/skills/matrixfounder/agentic-development/skill-creator.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 2 findings, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Agent Snooping · line 9 Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.Fix: Remove all code or instructions that list or read other skills' files or directories. Skills should operate independently; cross-skill access is a privilege escalation.
- medium Rogue Agent · line 246 Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.Fix: Remove any persistence mechanisms (cron jobs, startup scripts, state files). Skills should not maintain state across sessions without explicit user consent.
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 | $0.00045 | $0.05816 |
| Opus 5.5 | $0.00018 | $0.02326 |
| Sonnet 5.5 | $0.00009 | $0.01163 |
| Haiku 4.5 | $0.00005 | $0.00582 |
Grade A, and why
skill-creator 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 9d 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 — 437 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Skill Creator Guide
This skill provides the authoritative standard for creating and iteratively improving Agent Skills. It combines the Anthropic Skills Standard with our local architecture rules.
Core loop: Draft skill → Write test cases → Run evals (with-skill + baseline) → Review with user → Improve → Repeat.
Your job is to figure out where the user is in this process and help them progress. Maybe they want to create a skill from scratch — help narrow intent, write a draft, create tests, run them, iterate. Maybe they already have a draft — go straight to eval/iterate. Be flexible.
Red Flags (Anti-Rationalization)
STOP and READ THIS if you are thinking:
- "I'll skip the eval step, the skill looks fine" → WRONG. Run evals — untested skills fail silently in production.
- "I can write the whole skill without talking to the user" → WRONG. Capture Intent first — assumptions cause rewrites.
- "The description is descriptive enough" → WRONG. CSO triggers are mechanical. Follow the schema.
- "This skill is too simple for a script" → WRONG. If logic > 5 lines, text instructions fail 30% of the time. Use a script.
- "I'll skip the viewer and evaluate outputs myself" → WRONG. Generate the eval viewer BEFORE evaluating — get results in front of the human ASAP.
Rationalization Table
| Agent Excuse | Reality / Counter-Argument |
|---|---|
| "The skill worked on my one test prompt" | One prompt is the prompt you wrote the skill against. A skill is judged on prompts nobody wrote it for; that is what the eval set is. |
| "Baseline runs are a waste of tokens" | Without a baseline you cannot tell a skill that helps from a model that already knew. The comparison IS the result. |
| "I'll invent the eval queries from the SKILL.md" | Queries derived from the skill test the skill against itself. Derive them from how a user would actually ask, including the phrasings that must NOT trigger it. |
| "Validation is a formality, I'll run it at the end" | validate_skill.py fails on frontmatter and structure — the cheapest failures to fix and the most expensive to discover after the content is written. Run it first. |
"A relative --path is fine for init_skill.py" |
It is resolved with os.path.abspath() against your current working directory, so the same command lands somewhere different depending on where you ran it. Pass an absolute --path. |
What ships with it
37 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.
- agents/analyzer.md 3.4 KB
- agents/comparator.md 2.6 KB
- agents/grader.md 5.4 KB
- assets/eval_review.html 6.9 KB
- assets/SKILL_TEMPLATE.md 3.2 KB
- eval-viewer/generate_review.py 16 KB runs code
- eval-viewer/viewer.html 44 KB
- examples/SKILL_EXAMPLE_LEGACY_MIGRATOR.md 2.6 KB
- references/advanced-eval-patterns.md 8.4 KB
- references/default_parameters.md 2.8 KB
- references/eval_schemas.md 14 KB
- references/output-patterns.md 1.8 KB
- references/persuasion-principles.md 5.8 KB
- references/skill_design_patterns.md 2.4 KB
- references/testing-skills-with-subagents.md 12 KB
- references/workflows.md 818 B
- references/writing_skills_best_practices_anthropic.md 40 KB
- scripts/__init__.py 0 B runs code
- scripts/aggregate_benchmark.py 21 KB runs code
- scripts/generate_report.py 7.0 KB runs code
- scripts/improve_description.py 11 KB runs code
- scripts/init_skill.py 9.0 KB runs code
- scripts/package_skill.py 5.6 KB runs code
- scripts/run_eval.py 24 KB runs code
- scripts/run_loop.py 14 KB runs code
- scripts/skill_standards_default.yaml 1.9 KB
- scripts/skill_utils.py 24 KB runs code
- scripts/tests/__init__.py 0 B runs code
- scripts/tests/_stdout_pipe_child.py 2.9 KB runs code
- scripts/tests/fixtures/fake_cli_claude_code.py 3.9 KB runs code
- scripts/tests/test_eval_tooling.py 11 KB runs code
- scripts/tests/test_frontmatter_strict.py 3.3 KB runs code
- scripts/tests/test_human_channel.py 25 KB runs code
- scripts/tests/test_stdout_broken_pipe.py 4.2 KB runs code
- scripts/tests/test_trigger_detection.py 15 KB runs code
- scripts/validate_skill.py 25 KB runs code
- scripts/verify_pin.py 6.7 KB runs code
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.
- 9d ago Changed 7a75411e9fdf
- 13d ago Changed · +2 lines 05227ca3160c
- 1mo ago Changed · +26 lines 988c34e8c8d0
- 1mo ago First seen · 409 lines · 45 tokens per session scan A 88584bbfe5a6
skill-creator is a skill published in the GitHub repository MatrixFounder/Agentic-development (6 stars, last pushed yesterday), licensed Apache-2.0. It adds 45 tokens to every session and 5,816 once invoked, about $0.0002 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 2026-08-31.
Other skills, from other repositories
08-debug
Reproduce and fix a known bug, or find an unknown root cause by hypothesis validation. Use when the user wants to fix a bug, find why something breaks, or reopen a stuck investigation. Not for building a feature or reviewing a diff.
workflow
Perform any development work on the ambient-ui component. Full workflow instructions.
chimera-write-the-check-before-the-code
A criterion written after the implementation describes the implementation. Write the failing check first, watch it fail, then build.
Learn what defines effective BDD scenarios
Interactive guidance on writing complete, effective BDD scenarios for story-flow.
hatch3r-bug-fix
Step-by-step bug fix workflow. Diagnose root cause, implement minimal fix, write regression test. Use when fixing bugs, working on bug report issues, or when the user mentions a bug.
ap-policies
Attach a completion policy gate (shell check or judge-agent rubric) to a session or fan-in group so turn-end only passes when the gate is green, then optionally auto-commit pending human ack. Use when the user says "gate this session on tests passing", "attach a policy", "commit only if tests are green", "judge…