All skills & plugins / Claude Code official plugins

claude-security

Deep vulnerability scanning of your own code, run entirely inside your Claude Code session at a chosen effort tier, with every finding challenged before it is reported and the verification tally computed in code. Turns surviving findings into targeted patches, each verified by a panel of agents, that you apply when you choose.

Type: Plugin by Anthropic
Catalog: Claude Code official plugins
Source: anthropics/claude-plugins-official/plugins/claude-security (46 files scanned)
Install and details: Agent Skills Index entry
License: Proprietary (LICENSE file)
Signals: Pre-approved shellHooksBundled scriptsSpawns processes

Pre-approved tools

Shell (any command). Tools listed in allowed-tools (or an agent's tools) can be used without a permission prompt while the skill is active, depending on your agent and settings.

Kind	Tools
read/search	`ReadGlobGrepAskUserQuestion`
write/edit files	`WriteEdit`
other	`WorkflowWorkflow(claude-security:scan)TaskCreateTaskGetTaskListTaskUpdateTaskOutputTaskStop`
sub-agents	`Agent(claude-security:scan-inventory, claude-security:scan-researcher, claude-security:scan-verifier, claude-security:scan-loader, claude-security:scan-redactor, claude-security:patch-generator, claude-security:patch-verifier, claude-security:explore)Agent(claude-security:explore)`
shell, scoped	`Bash(date )Bash(ls )Bash(wc )Bash(mkdir -p )Bash(git )Bash(GIT_CONFIG_GLOBAL=/dev/null GIT_TERMINAL_PROMPT=0 git )Bash(find . -maxdepth 1 -type d -name "CLAUDE-SECURITY-2")Bash(python3 "${CLAUDE_PLUGIN_ROOT}/scripts/render_report.py" )Bash(python3 "${CLAUDE_PLUGIN_ROOT}/scripts/save_result.py" )Bash(python3 "${CLAUDE_PLUGIN_ROOT}/scripts/write_scan_meta.py" )Bash(bash "${CLAUDE_PLUGIN_ROOT}/scripts/keep-waiting.sh" )Bash(python3 "${CLAUDE_PLUGIN_ROOT}/scripts/patch_artifacts.py" )Bash(sleep )Bash(GIT_TERMINAL_PROMPT=0 git )`
shell, any command	`Bash`

Declared in claude-security/SKILL.md, agents/patch-verifier.md, agents/scan-loader.md, agents/explore.md, agents/claude-security.md, agents/scan-researcher.md (first 60 of 71 shown).

Hooks

Hooks run automatically on these agent events, without you invoking the skill: PermissionRequest, PostToolUse, PostToolUseFailure, UserPromptExpansion.

Event	Matcher	Runs
UserPromptExpansion	^claude-security:claude-security$	`sh "${CLAUDE_PLUGIN_ROOT}/hooks/hooks.sh" banner` hooks/hooks.json
PostToolUse	Bash	`sh "${CLAUDE_PLUGIN_ROOT}/hooks/hooks.sh" metrics \|\| true` hooks/hooks.json
PostToolUse	Bash	`sh "${CLAUDE_PLUGIN_ROOT}/hooks/hooks.sh" tip push \|\| true` hooks/hooks.json
PostToolUse	Bash	`sh "${CLAUDE_PLUGIN_ROOT}/hooks/hooks.sh" tip pr \|\| true` hooks/hooks.json
PostToolUseFailure	Bash	`sh "${CLAUDE_PLUGIN_ROOT}/hooks/hooks.sh" metrics \|\| true` hooks/hooks.json
PermissionRequest	AskUserQuestion	`sh "${CLAUDE_PLUGIN_ROOT}/hooks/hooks.sh" unanswered \|\| true` hooks/hooks.json

MCP servers and apps

None.

Bundled code

22 code files (JavaScript, Python, Shell):

Network calls

No HTTP or socket call pattern found in the code.

Starts other programs

4 files. First match per file:

plugins/claude-security/scripts/patch_artifacts.py:535
```
out = subprocess.run(
```
plugins/claude-security/scripts/lib/git.py:15
```
out = subprocess.run(
```
plugins/claude-security/hooks/hooks.py:322
```
out = subprocess.run(
```

plugins/claude-security/workflows/scan.js:50

;for(const n of e){const e=(/\.([^./]+)$/.exec(n)||[])[1];(void 0===e?Ze.has(n.split("/").pop().toUpperCase()):Qe.has(e))||t.add(Xe.get(e)||

Credentials referenced

No key, token or secret environment variable found in code or MCP config.

Static scan of commit 41f619d on 2026-10-27. Signals are patterns found in the files, not a verdict: a network call can be the whole point of a skill. Read the linked lines before you decide. How the scan works.