MCP Server Security Audit
awarexone/Agentic-Bug-Hunter
Audits MCP servers and their client configs for tool poisoning, prompt injection, over-privileged tools, injection bugs, secret leaks and missing approval gates.
Add security coverage to Guardana the way this repository requires — as a rule, evaluator or target, never by patching the engine — with the fixtures, the framework mapping and the documentation…
Install Add A Rule
$ npx skills add guardana/guardana --skill add-a-rule -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install guardana/guardana add-a-rule --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/guardana/guardana.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude/skills/add-a-rule .claude/skills/add-a-rule && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "add-a-rule" agent skill from https://github.com/guardana/guardana/tree/main/.claude/skills/add-a-rule into .claude/skills/add-a-rule/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "add-a-rule", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/guardana/guardana/tree/main/.claude/skills/add-a-ruleType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add guardana/guardana --skill add-a-rule -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install guardana/guardana add-a-rule --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/guardana/guardana.git skills-src && mkdir -p .agents/skills && cp -r skills-src/.claude/skills/add-a-rule .agents/skills/add-a-rule && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "add-a-rule" agent skill from https://github.com/guardana/guardana/tree/main/.claude/skills/add-a-rule into .agents/skills/add-a-rule/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "add-a-rule", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add guardana/guardana --skill add-a-rule -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install guardana/guardana add-a-rule --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/guardana/guardana.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/.claude/skills/add-a-rule .cursor/skills/add-a-rule && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "add-a-rule" agent skill from https://github.com/guardana/guardana/tree/main/.claude/skills/add-a-rule into .cursor/skills/add-a-rule/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "add-a-rule", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/guardana/guardana.git --path .claude/skills/add-a-rule--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add guardana/guardana --skill add-a-rule -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install guardana/guardana add-a-rule --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/guardana/guardana.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/.claude/skills/add-a-rule .gemini/skills/add-a-rule && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "add-a-rule" agent skill from https://github.com/guardana/guardana/tree/main/.claude/skills/add-a-rule into .gemini/skills/add-a-rule/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "add-a-rule", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install guardana/guardana add-a-ruleInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add guardana/guardana --skill add-a-rule -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/guardana/guardana.git skills-src && mkdir -p .github/skills && cp -r skills-src/.claude/skills/add-a-rule .github/skills/add-a-rule && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "add-a-rule" agent skill from https://github.com/guardana/guardana/tree/main/.claude/skills/add-a-rule into .github/skills/add-a-rule/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "add-a-rule", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add guardana/guardana --skill add-a-rule -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install guardana/guardana add-a-rule --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/guardana/guardana.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/.claude/skills/add-a-rule .opencode/skills/add-a-rule && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "add-a-rule" agent skill from https://github.com/guardana/guardana/tree/main/.claude/skills/add-a-rule into .opencode/skills/add-a-rule/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "add-a-rule", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
add-a-rule- GitHub stars
- 131
- Token cost
- ~1.1k tokens
- SKILL.md length
- 614 words
- Files
- 1
- Skills in repo
- 13
- Repo updated
- First seen
- Licence
- Apache-2.0
At a glance
Add security coverage to Guardana the way this repository requires — as a rule, evaluator or target, never by patching the engine — with the fixtures, the framework mapping and the documentation…
- Asked to add a check
- SKILL.md covers Pick the path, Non-negotiables, Then the documentation, in the… and Verify it the way the project…
- Calls uv
- Cover a new threat
What it does
Add A Rule is an agent skill from guardana/guardana. Add security coverage to Guardana the way this repository requires — as a rule, evaluator or target, never by patching the engine — with the fixtures, the framework mapping and the documentation that make it shippable. Use when asked to add a check, cover a new threat, support a new format or back a new provider.
Its SKILL.md is about 1.1k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.
It sits in Security, covering Prompt injection and agent security. It works with Model Context Protocol and Python. The repository describes itself as: Open-source AI security verification for model artifacts, live endpoints, MCP servers, and recorded agent traces. Reproducible evidence for release decisions. The licence is Apache-2.0.
When your agent uses it
- Asked to add a check
- Cover a new threat
- Support a new format
- Back a new provider
Example prompts
- “/add-a-rule”
Requirements
- Python 3
What it can do on your machine
Read from SKILL.md and the folder at commit 616c63b. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
Shell commands in SKILL.md call:
uv
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md. Its commands use uv, which can reach the network depending on how they are called.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
Add A Rule loads about 1.1k tokens when it runs. Until then it costs about 81 tokens; SKILL.md has 614 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
SKILL.md
The full file from guardana/guardana at commit 616c63b, republished under its Apache-2.0 licence (© guardana). 614 words, ~1,106 tokens.
Download SKILL.mdSave it as
.claude/skills/add-a-rule/SKILL.md (or your agent's skills folder).- name
- add-a-rule
- description
- Add security coverage to Guardana the way this repository requires — as a rule, evaluator or target, never by patching the engine — with the fixtures, the framework mapping and the documentation that make it shippable. Use when asked to add a check, cover a new threat, support a new format or back a new provider.
Adding coverage
The engine knows almost nothing about specific threats. It knows how to discover
rules, run them against targets, and evaluate outcomes. All domain knowledge
lives in rules, evaluators and targets — never in the engine. A change that
teaches guardana-core about a threat, a vendor, a file format or a regulation
is the wrong change, however small.
Pick the path
YAML is the default, for anything expressible as "send this prompt, grade it
with this evaluator". No code. Drop it in packages/guardana-rules/src/guardana/ rules/catalog/; uv run guardana new-rule acme.prompt.my_check scaffolds one.
steps: instead of prompts: makes it a multi-turn ScenarioRule.
A Python plugin is for logic YAML cannot express — a parser, a stateful
probe, an artifact format. Same Rule contract, registered through the
guardana.rules entry point.
An evaluator when the judgement is what is new, not the stimulus. One short
file, registered through guardana.evaluators. A rule names it by string, so
swapping graders never touches a rule.
A target when the thing under test is new. Declare capabilities() and
implement the matching protocol from guardana.core.target.protocols — both
halves, then prove it with guardana.testing.assert_target_conforms.
Non-negotiables
A framework mapping, or it does not ship. Every built-in security rule maps to
OWASP LLM / OWASP ASI / MITRE ATLAS / NIST. The mapping is what makes a finding
answerable in somebody else's audit. Use the full reference form (LLM07:2025),
never a bare id. A team's own quality criteria (suites, local checks) need no
public mapping, and a built-in one is never invented to fill the field.
A positive fixture and a negative one. The positive proves the rule fires;
the negative proves it stays quiet. Dynamic rules get both in three lines with
guardana.core.testing's scripted transports and no network. A rule with only a
positive fixture is a rule nobody has shown to be quiet, and guardana rule test
reports it as indeterminate rather than green — truthfully.
Never a confident all-clear on something unexamined. If the rule cannot run —
no canary was planted, a judge's reply is unparseable, the model returned no text
— the verdict is inconclusive or a finding. In this codebase, silence is never
spelled pass. No linter can catch this; only an adversarial reader can.
Cost grows with the target, not with the rule count. A new rule must not add
a tree walk, a re-read or a re-parse of something already read this run. Ask
through target.python_source(path) and target.iter_files(suffixes); both
cache.
Show full SKILL.md (216 more words)Show less
Declare impact and cost. impact, destructive and estimated_requests are
how a policy selects and a budget bounds. estimated_requests is an upper bound:
spending less is fine, spending more is a defect, and a gate measures every
shipped rule against its own declaration.
Namespace it. guardana.* is reserved for built-ins and is now enforced —
an installed distribution registering one is refused at load time.
Record what you measured, if you graded something. A rule that sends a prompt
and judges a reply calls ctx.record(from_verdict(...)) for every case, passes
included. Without the passes there is no denominator. A rule that only reads a
file records nothing — "I looked and found nothing" is not a measurement.
Then the documentation, in the same commit
CHANGELOG.md (why, not only what), FEATURES.md if the capability surface
moved, the relevant docs/ page, and uv run python scripts/generate_docs.py —
never edit docs/generated/ by hand. The landing page states a rule count; a
test pins it to the registry, so it will tell you.
Verify it the way the project verifies things
Run the gate (gate skill), and then run the documented command against a
real target and read what it wrote. A fake OpenAI-compatible endpoint is three
lines of http.server. More real defects in this repository have been found that
way than by any test.
© guardana, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
Just SKILL.md in .claude/skills/add-a-rule of guardana/guardana.
Open the folder on GitHubat commit 616c63b
Compare with similar skills
Add A Rule next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Add A Rule this skillguardana/guardana | 131 | — | ~1.1k | Automated safety check: Pass | Apache-2.0 | |
| MCP Server Security Auditawarexone/Agentic-Bug-Hunter | 5.3k | — | ~1.9k | Automated safety check: Warn | MIT | |
| Skill InspectorNVIDIA/SkillSpector | 20k | — | ~1.8k | Automated safety check: Pass | Apache-2.0 | |
| Vulners API Python SDKvulnersCom/api | 376 | — | ~2.3k | Automated safety check: Pass | MIT | |
| Forensifyalexgreensh/repo-forensics | 190 | — | ~2.5k | Automated safety check: Notes | Custom licence | |
| Skylos Securityduriantaco/skylos | 844 | — | ~545 | Automated safety check: Pass | Apache-2.0 |
Similar skills
- Official
Skill Inspector
NVIDIA/SkillSpector
Decides whether an agent skill is safe to install by combining a SkillSpector static scan with the agent's own source review, ending in APPROVE, CAUTION or REJECT.
Vulners API Python SDK
vulnersCom/api
A skill your agent uses when modifying, testing, documenting, or reviewing the Vulners Python SDK.
Forensify
alexgreensh/repo-forensics
Cross-agent self-inspection of your AI-agent stack. An agent skill from alexgreensh/repo-forensics.
Skylos Security
duriantaco/skylos
Investigate and harden Skylos security behavior. An agent skill from duriantaco/skylos.
Slowmist Agent Security
slowmist/slowmist-agent-security
Comprehensive security review framework for AI agents. An agent skill from slowmist/slowmist-agent-security.
More from guardana/guardana
All 13 skills in this repoShip
guardana/guardana
Commit and push a finished change the way this repo requires — explicit paths, one commit per logical change, no attribution, the five documentation places answered, the site checks before a push (a…
Auto
guardana/guardana
Autonomous, non-interactive run of the whole lifecycle for one development task — size, plan, build, gate, review, fix, commit — without stopping for questions.
Docs
guardana/guardana
Documentation work in this repo — the five places a user-visible change must answer, the page conventions the site build enforces, the tests that pin prose to the registry, and a simplification pass…
False Green Audit
guardana/guardana
Hunt for the failure this project exists to prevent — code that compiles, types, tests green, and quietly reports "all clear" about something it never examined.
Gate
guardana/guardana
Run this project's verification — the full local CI mirror (ruff, mypy, import contract, pytest with PostgreSQL, coverage floors, dogfood, generated docs and site, the isolated example suites, the…
Plan
guardana/guardana
Turn a development task into one work file in .work/ — goal, decisions, lanes with exact files and verification, done-criteria.
Related
Works with
Categories
Questions about Add A Rule
What does Add A Rule do?
Add security coverage to Guardana the way this repository requires — as a rule, evaluator or target, never by patching the engine — with the fixtures, the framework mapping and the documentation…. Add A Rule is an agent skill from guardana/guardana. Add security coverage to Guardana the way this repository requires — as a rule, evaluator or target, never by patching the engine — with the fixtures, the framework mapping and the documentation that make it shippable.
When should I use Add A Rule?
Add A Rule fits situations like: asked to add a check; cover a new threat; support a new format; back a new provider.
How do I install Add A Rule in Claude Code?
Run `npx skills add guardana/guardana --skill add-a-rule -a claude-code`. Or copy the skill folder (.claude/skills/add-a-rule in guardana/guardana) into .claude/skills/add-a-rule in your project. Claude Code loads it when a task matches its description.
How do I install Add A Rule in Codex?
Run `npx skills add guardana/guardana --skill add-a-rule -a codex`. Or copy the skill folder (.claude/skills/add-a-rule in guardana/guardana) into .agents/skills/add-a-rule in your project. Codex loads it when a task matches its description.
Can I use Add A Rule in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add guardana/guardana --skill add-a-rule -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/add-a-rule, .gemini/skills/add-a-rule, .github/skills/add-a-rule and .opencode/skills/add-a-rule in your project.
What does Add A Rule need to run?
Going by SKILL.md and its folder, Add A Rule needs the command-line tools its instructions call (uv). Our summary lists: Python 3.
Does Add A Rule access the network?
SKILL.md contains no URLs. Its commands use uv, which can reach the network depending on how they are called. This is read from the text; nothing was executed.
Is Add A Rule safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
What licence does Add A Rule use?
Add A Rule is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Add A Rule use?
About 1.1k tokens (SKILL.md is roughly 4.4k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
What are the alternatives to Add A Rule?
Skills that share tags, products or a category with Add A Rule: MCP Server Security Audit (awarexone/Agentic-Bug-Hunter, 5.3k stars), Skill Inspector (NVIDIA/SkillSpector, 20k stars), Vulners API Python SDK (vulnersCom/api, 376 stars) and Forensify (alexgreensh/repo-forensics, 190 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Add A Rule?
guardana (a GitHub organization) maintains it in guardana/guardana, which has 131 GitHub stars. The repository holds 13 skills in this directory. The repository was last updated on October 10, 2026.
Source: guardana/guardana on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.