Agento-patronum patronum-suggest

Suggest protection patterns based on project context. Invoke automatically when user mentions a new tech stack, cloud provider, or sensitive tooling. Also invoke when user asks what to protect.

install
source · Clone the upstream repo
git clone https://github.com/emaarco/agento-patronum
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/emaarco/agento-patronum "$T" && mkdir -p ~/.claude/skills && cp -r "$T/skills/patronum-suggest" ~/.claude/skills/emaarco-agento-patronum-patronum-suggest && rm -rf "$T"
manifest: skills/patronum-suggest/SKILL.md
source content

Skill: patronum-suggest

Analyze the current project and suggest relevant protection patterns.

Steps

1. Detect tech stack

Use

Glob
to list files in the project. Do NOT read file contents — infer the stack from file names and paths alone.

Signals to look for:

  • package.json
    → Node.js / npm
  • go.mod
    → Go
  • requirements.txt
    ,
    pyproject.toml
    ,
    Pipfile
    → Python
  • Gemfile
    → Ruby
  • Cargo.toml
    → Rust
  • **/*.tf
    → Terraform
  • docker-compose.yml
    ,
    Dockerfile
    → Docker
  • .gcloud/
    ,
    .azure/
    directories → GCP / Azure
  • .github/workflows/*.yml
    ,
    .gitlab-ci.yml
    → CI/CD
  • pom.xml
    ,
    build.gradle
    → Java / Gradle / Maven
  • *.kubeconfig
    ,
    kubeconfig
    → Kubernetes

2. Research sensitive files

Use

WebSearch
to find known sensitive files, credential paths, and secret locations for the detected technologies. Search for patterns like:

  • "[technology] sensitive files credentials path"
  • "[cloud provider] local config files secrets"

This ensures suggestions cover technology-specific risks beyond the hardcoded list.

3. Check current protections

Run:

node "${CLAUDE_PLUGIN_ROOT}/scripts/management/patronum-list.js"
to see what is already protected.

4. Build suggestions

Based on detected stack and web research, suggest patterns that are NOT yet protected. Common suggestions include:

  • Terraform:
    **/*.tfvars
    ,
    **/*.tfstate
    ,
    **/.terraform/environment
  • GCP:
    ~/.config/gcloud/credentials.db
    ,
    **/service-account*.json
  • Azure:
    ~/.azure/accessTokens.json
    ,
    ~/.azure/msal_token_cache.json
  • Ruby:
    ~/.gem/credentials
  • Gradle/Maven:
    ~/.gradle/gradle.properties
    ,
    ~/.m2/settings.xml
  • Kubernetes:
    **/kubeconfig
    ,
    **/*.kubeconfig
  • Vault:
    ~/.vault-token
  • GPG:
    ~/.gnupg/*

Include any additional patterns discovered via web search.

5. Confirm with user

Use

AskUserQuestion
to present the suggestions as a formatted list with reasons. Let the user select which patterns to add.

6. Add confirmed patterns

For each confirmed pattern, run:

node "${CLAUDE_PLUGIN_ROOT}/scripts/management/patronum-add.js" "<pattern>" --reason "<reason>"

Present the final updated protection list as a markdown table.