用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt;运行前请先检查来源。
npx skills add https://github.com/duclm1x1/Dive-Ai --skill clawsec-feed命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地?可下载 SkillsMP 当前能够提供的文件。
Persistent memory system for AI agents following Model Context Protocol (MCP). Use for storing long-term memories across sessions, semantic search of past knowledge, building knowledge graphs, auto-injecting context, deduplicating memories, syncing to cloud storage. Essential for agents that need to remember decisions, solutions, preferences, and learned patterns over time.
Persistent memory system for AI agents following Model Context Protocol (MCP). Use for storing long-term memories across sessions, semantic search of past knowledge, building knowledge graphs, auto-injecting context, deduplicating memories, syncing to cloud storage. Essential for agents that need to remember decisions, solutions, preferences, and learned patterns over time.
Master REST and GraphQL API design principles to build intuitive, scalable, and maintainable APIs that delight developers. Use when designing new APIs, reviewing API specifications, or establishing API design standards.
基于 SOC 职业分类
正在显示 SKILL.md
| name | clawsec-feed |
| version | 0.0.4 |
| description | Security advisory feed with automated NVD CVE polling for OpenClaw-related vulnerabilities. Updated daily. |
| homepage | https://clawsec.prompt.security |
| metadata | {"openclaw":{"emoji":"📡","category":"security"}} |
| clawdis | {"emoji":"📡","requires":{"bins":["curl","jq"]}} |
Security advisory feed monitoring for AI agents. Subscribe to community-driven threat intelligence and stay informed about emerging threats.
This feed is automatically updated daily with CVEs related to OpenClaw, clawdbot, and Moltbot from the NIST National Vulnerability Database (NVD).
An open source project by Prompt Security
You can get clawsec-feed in two ways:
If you've installed clawsec-suite, you may already have this!
ClawSec-feed is bundled alongside ClawSec Suite to provide crucial CVE and threat intelligence information. When you install the suite, if you don't already have the feed installed, it will be deployed from the bundled copy.
Advantages:
~/.openclaw/skills/clawsec-feed/Install clawsec-feed independently without the full suite.
When to use standalone:
Advantages:
Continue below for standalone installation instructions.
Installation Steps: Agents should review scripts before execution!
# Get latest clawsec-feed release tag
LATEST_TAG=$(curl -sSL https://api.github.com/repos/prompt-security/ClawSec/releases | \
jq -r '[.[] | select(.tag_name | startswith("clawsec-feed-v"))][0].tag_name')
# Download SKILL.md from latest release
SKILL_URL="https://github.com/prompt-security/clawsec/releases/download/$LATEST_TAG/SKILL.md"
curl -sSL --fail --show-error "$SKILL_URL"
Once you have this skill file, proceed to Deploy ClawSec Feed below.
Installation steps:
mkdir -p ~/.openclaw/skills/clawsec-feed
# Save this SKILL.md as SKILL.md in the directory above
# Get latest release tag with retry logic
LATEST_TAG=$(curl -sSL --retry 3 --retry-delay 1 \
https://api.github.com/repos/prompt-security/ClawSec/releases | \
jq -r '[.[] | select(.tag_name | startswith("clawsec-feed-v"))][0].tag_name')
BASE_URL="https://github.com/prompt-security/clawsec/releases/download/$LATEST_TAG"
INSTALL_DIR="${CLAWSEC_INSTALL_DIR:-$HOME/.openclaw/skills/clawsec-feed}"
TEMP_DIR=$(mktemp -d)
trap "rm -rf '$TEMP_DIR'" EXIT
# Download checksums.json (REQUIRED for integrity verification)
echo "Downloading checksums..."
if ! curl -sSL --fail --show-error --retry 3 --retry-delay 1 \
"$BASE_URL/checksums.json" -o "$TEMP_DIR/checksums.json"; then
echo "ERROR: Failed to download checksums.json"
exit 1
fi
# Validate checksums.json structure
if ! jq -e '.skill and .version and .files' "$TEMP_DIR/checksums.json" >/dev/null 2>&1; then
echo "ERROR: Invalid checksums.json structure"
exit 1
fi
# PRIMARY: Try .skill artifact
echo "Attempting .skill artifact installation..."
if curl -sSL --fail --show-error --retry 3 --retry-delay 1 \
"$BASE_URL/clawsec-feed.skill" -o "$TEMP_DIR/clawsec-feed.skill" 2>/dev/null;
ARTIFACT_SIZE=$( -c%s 2>/dev/null || -f%z )
MAX_SIZE=$(( * * ))
[ -gt ];
unzip -l | grep -qE ;
1
FILE_COUNT=$(unzip -l | grep -c || 0)
[ -gt 100 ];
1
unzip -q -d
[ ! -f ];
1
CHECKSUM_FAILED=0
file $(jq -r );
EXPECTED=$(jq -r --arg f )
FILE_PATH=$(jq -r --arg f )
[ -f ];
ACTUAL=$(shasum -a 256 | -d -f1)
[ -f ];
ACTUAL=$(shasum -a 256 | -d -f1)
CHECKSUM_FAILED=1
[ != ];
CHECKSUM_FAILED=1
[ -eq 0 ];
[ -f ];
FEED_FILE=
[ -f ];
FEED_FILE=
1
! jq -e >/dev/null 2>&1;
1
-p
-r /*
600
find - f ! -name - 644 {} \;
0
-p
DOWNLOAD_FAILED=0
file $(jq -r );
FILE_URL=$(jq -r --arg f )
EXPECTED=$(jq -r --arg f )
! curl -sSL --fail --show-error --retry 3 --retry-delay 1 \
-o ;
DOWNLOAD_FAILED=1
ACTUAL=$(shasum -a 256 | -d -f1)
[ != ];
DOWNLOAD_FAILED=1
[ -eq 1 ];
1
! jq -e >/dev/null 2>&1;
1
! jq -e >/dev/null 2>&1;
1
-p
/*
600
find - f ! -name - 644 {} \;
Add ClawSec Feed to your heartbeat routine:
## ClawSec Feed (every heartbeat)
1. Check advisory feed for new security alerts
2. Cross-reference with installed skills
3. Notify user if action is required
Tell your user:
ClawSec Feed deployed successfully. I will now:
- Monitor the community advisory feed for threats
- Alert you to any security issues affecting your environment
- Keep you informed of emerging attack patterns
Subscribes to the community advisory feed for:
When a relevant advisory is published, your agent will notify you.
# Use environment variable if set, otherwise use raw GitHub feed (always up-to-date)
DEFAULT_FEED_URL="https://raw.githubusercontent.com/prompt-security/ClawSec/main/advisories/feed.json"
FEED_URL="${CLAWSEC_FEED_URL:-$DEFAULT_FEED_URL}"
# Fetch with error handling and retry logic
curl -sSL --fail --show-error --retry 3 --retry-delay 1 "$FEED_URL"
Feed structure:
{
"version": "1.0",
"updated": "2026-02-02T12:00:00Z",
"advisories": [
{
"id": "GA-2026-001",
"severity": "critical",
"type": "malicious_skill",
"title": "Malicious data exfiltration in skill 'helper-plus'",
"description": "Skill sends user data to external server",
"affected": ["helper-plus@1.0.0", "helper-plus@1.0.1"],
"action": "Remove immediately",
"published": "2026-02-01T10:00:00Z"
}
]
}
# Use environment variable if set, otherwise use raw GitHub feed (always up-to-date)
DEFAULT_FEED_URL="https://raw.githubusercontent.com/prompt-security/ClawSec/main/advisories/feed.json"
FEED_URL="${CLAWSEC_FEED_URL:-$DEFAULT_FEED_URL}"
TEMP_FEED=$(mktemp)
trap "rm -f '$TEMP_FEED'" EXIT
if ! curl -sSL --fail --show-error --retry 3 --retry-delay 1 "$FEED_URL" -o "$TEMP_FEED"; then
echo "Error: Failed to fetch advisory feed"
exit 1
fi
# Validate JSON before parsing
if ! jq empty "$TEMP_FEED" 2>/dev/null; then
echo "Error: Invalid JSON in feed"
exit 1
fi
FEED=$(cat "$TEMP_FEED")
# Get advisory count with error handling
COUNT=$(echo "$FEED" | jq '.advisories | length')
if [ $? -ne 0 ]; then
echo "Error: Failed to parse advisories"
exit 1
fi
echo "Advisory count: $COUNT"
# Parse critical advisories with jq error handling
CRITICAL=$(echo "$FEED" | jq '.advisories[] | select(.severity == "critical")')
if [ $? -ne 0 ]; then
echo "Error: Failed to filter critical advisories"
exit 1
fi
echo "$CRITICAL"
# Use UTC timezone for consistent date handling
WEEK_AGO=$(TZ=UTC date -v-7d +%Y-%m-%dT00:00:00Z 2>/dev/null || TZ=UTC date -d '7 days ago' +%Y-%m-%dT00:00:00Z)
RECENT=$(echo "$FEED" | jq --arg since "$WEEK_AGO" '.advisories[] | select(.published > $since)')
if [ $? -ne 0 ]; then
echo "Error: Failed to filter recent advisories"
exit 1
fi
echo "$RECENT"
Check if any of your installed skills are affected by advisories:
# List your installed skills (adjust path for your platform)
INSTALL_DIR="${CLAWSEC_INSTALL_DIR:-$HOME/.openclaw/skills}"
# Use environment variable if set, otherwise use raw GitHub feed (always up-to-date)
DEFAULT_FEED_URL="https://raw.githubusercontent.com/prompt-security/ClawSec/main/advisories/feed.json"
FEED_URL="${CLAWSEC_FEED_URL:-$DEFAULT_FEED_URL}"
TEMP_FEED=$(mktemp)
trap "rm -f '$TEMP_FEED'" EXIT
if ! curl -sSL --fail --show-error --retry 3 --retry-delay 1 "$FEED_URL" -o "$TEMP_FEED"; then
echo "Error: Failed to fetch advisory feed"
exit 1
fi
# Validate and parse feed
if ! jq empty "$TEMP_FEED" 2>/dev/null; then
echo "Error: Invalid JSON in feed"
exit 1
fi
FEED=$(cat "$TEMP_FEED")
AFFECTED=$(echo "$FEED" | jq -r '.advisories[].affected[]?' 2>/dev/null | sort -u)
if [ $? -ne 0 ]; then
echo "Error: Failed to parse affected skills from feed"
exit 1
fi
# Safely validate all installed skills before processing
VALIDATED_SKILLS=()
IFS= -r -d skill_path;
skill=$( )
[[ =~ ^[a-zA-Z0-9_-]+$ ]];
VALIDATED_SKILLS+=()
>&2
< <(find -mindepth 1 -maxdepth 1 - d -print0 2>/dev/null)
skill ;
| grep -qF ;
| jq --arg s
If you find affected skills:
| Type | Description |
|---|---|
malicious_skill | Skill identified as intentionally harmful |
vulnerable_skill | Skill with security vulnerabilities |
prompt_injection | Known prompt injection pattern |
attack_pattern | Observed attack technique |
best_practice | Security recommendation |
| Severity | Action Required |
|---|---|
critical | Notify user immediately, take action |
high | Notify user soon, plan remediation |
medium | Notify at next interaction |
low | Log for reference |
Notify Immediately (Critical):
Notify Soon (High):
Notify at Next Interaction (Medium):
Log Only (Low/Info):
📡 ClawSec Feed: 2 new advisories since last check
CRITICAL - GA-2026-015: Malicious prompt pattern "ignore-all"
→ Detected prompt injection technique. Update your system prompt defenses.
HIGH - GA-2026-016: Vulnerable skill "data-helper" v1.2.0
→ You have this installed! Recommended action: Update to v1.2.1 or remove.
FEED_OK - Advisory feed checked, no new alerts. 📡
Track the last feed check to identify new advisories:
{
"schema_version": "1.0",
"last_feed_check": "2026-02-02T15:00:00Z",
"last_feed_updated": "2026-02-02T12:00:00Z",
"known_advisories": ["GA-2026-001", "GA-2026-002"]
}
Save to: ~/.openclaw/clawsec-feed-state.json
STATE_FILE="$HOME/.openclaw/clawsec-feed-state.json"
# Create state file with secure permissions if it doesn't exist
if [ ! -f "$STATE_FILE" ]; then
echo '{"schema_version":"1.0","last_feed_check":null,"last_feed_updated":null,"known_advisories":[]}' > "$STATE_FILE"
chmod 600 "$STATE_FILE"
fi
# Validate state file before reading
if ! jq -e '.schema_version' "$STATE_FILE" >/dev/null 2>&1; then
echo "Warning: State file corrupted or invalid schema. Creating backup and resetting."
cp "$STATE_FILE" "${STATE_FILE}.bak.$(TZ=UTC date +%Y%m%d%H%M%S)"
echo '{"schema_version":"1.0","last_feed_check":null,"last_feed_updated":null,"known_advisories":[]}' > "$STATE_FILE"
chmod 600 "$STATE_FILE"
fi
# Check for major version compatibility
SCHEMA_VER=$(jq -r '.schema_version // "0"' "$STATE_FILE")
if [[ "${SCHEMA_VER%%.*}" != "1" ]]; then
echo "Warning: State file schema version may not be compatible with this version"
TEMP_STATE=$()
jq --arg t > ;
600
-f
Important: To avoid excessive requests to the feed server, follow these guidelines:
| Check Type | Recommended Interval | Minimum Interval |
|---|---|---|
| Heartbeat check | Every 15-30 minutes | 5 minutes |
| Full feed refresh | Every 1-4 hours | 30 minutes |
| Cross-reference scan | Once per session | 5 minutes |
# Check if enough time has passed since last check
STATE_FILE="$HOME/.openclaw/clawsec-feed-state.json"
MIN_INTERVAL_SECONDS=300 # 5 minutes
LAST_CHECK=$(jq -r '.last_feed_check // "1970-01-01T00:00:00Z"' "$STATE_FILE" 2>/dev/null)
LAST_EPOCH=$(TZ=UTC date -j -f "%Y-%m-%dT%H:%M:%SZ" "$LAST_CHECK" +%s 2>/dev/null || date -d "$LAST_CHECK" +%s 2>/dev/null || echo 0)
NOW_EPOCH=$(TZ=UTC date +%s)
if [ $((NOW_EPOCH - LAST_EPOCH)) -lt $MIN_INTERVAL_SECONDS ]; then
echo "Rate limit: Last check was less than 5 minutes ago. Skipping."
exit 0
fi
| Variable | Description | Default |
|---|---|---|
CLAWSEC_FEED_URL | Custom advisory feed URL | Raw GitHub (main branch) |
CLAWSEC_INSTALL_DIR | Installation directory | ~/.openclaw/skills/clawsec-feed |
Check for and install newer versions:
# Check current installed version
INSTALL_DIR="${CLAWSEC_INSTALL_DIR:-$HOME/.openclaw/skills/clawsec-feed}"
CURRENT_VERSION=$(jq -r '.version' "$INSTALL_DIR/skill.json" 2>/dev/null || echo "unknown")
echo "Installed version: $CURRENT_VERSION"
# Check latest available version
LATEST_URL="https://api.github.com/repos/prompt-security/ClawSec/releases"
LATEST_VERSION=$(curl -sSL --fail --show-error --retry 3 --retry-delay 1 "$LATEST_URL" 2>/dev/null | \
jq -r '[.[] | select(.tag_name | startswith("clawsec-feed-v"))][0].tag_name // empty' | \
sed 's/clawsec-feed-v//')
if [ -z "$LATEST_VERSION" ]; then
echo "Warning: Could not determine latest version"
else
echo "Latest version: $LATEST_VERSION"
if [ "$CURRENT_VERSION" != "$LATEST_VERSION" ]; then
echo "Update available! Run the deployment steps with the new version."
else
echo "You are running the latest version."
fi
fi
Bootstrap Trust Problem: The initial download of this skill cannot be verified by the skill itself. To establish trust:
https://clawsec.prompt.securitychecksums.json:# After downloading SKILL.md, verify its integrity
EXPECTED_HASH="<hash-from-checksums.json>"
ACTUAL_HASH=$(shasum -a 256 SKILL.md | cut -d' ' -f1)
if [ "$EXPECTED_HASH" != "$ACTUAL_HASH" ]; then
echo "ERROR: Skill file integrity check failed!"
echo "This file may have been tampered with. Do not proceed."
exit 1
fi
Note: For maximum security, verify checksums.json via a separate trusted channel (e.g., direct from GitHub release page UI, not via curl).
MIT License - See repository for details.
Built with 📡 by the Prompt Security team and the agent community.