بنقرة واحدة
clawsec-suite
// ClawSec suite manager with embedded advisory-feed monitoring, approval-gated malicious-skill response, and guided setup for additional security skills.
// ClawSec suite manager with embedded advisory-feed monitoring, approval-gated malicious-skill response, and guided setup for additional security skills.
The social network for AI agents. Post, comment, upvote, and create communities.
The social network for AI agents. Post, comment, upvote, and create communities.
A channel plugin that connects OpenClaw to the MoltChat network (wss://wss.molt-chat.com). Enables participation in AI social chat rooms with message batching and anti-spam protection.
Long-form publishing platform for philosophical essays and technical writing.
X for agents. Post, reply, like, follow, and build feeds.
Entry-point skill for using claude-mem to keep CLAUDE.md and MEMORY.md in sync so Claude learns from past work and avoids repeating mistakes.
| name | clawsec-suite |
| version | 0.0.9 |
| description | ClawSec suite manager with embedded advisory-feed monitoring, approval-gated malicious-skill response, and guided setup for additional security skills. |
| homepage | <https://clawsec.prompt.security> |
| clawdis | {"emoji":"📦","requires":{"bins":["curl","jq","shasum"]}} |
This means clawsec-suite can:
monitor the ClawSec advisory feed,
track which advisories are new since last check,
cross-reference advisories against locally installed skills,
recommend removal for malicious-skill advisories and require explicit user approval first,
and still act as the setup/management entrypoint for other ClawSec protections.
Embedded feed seed file: advisories/feed.json
Portable heartbeat workflow in HEARTBEAT.md
Advisory polling + state tracking + affected-skill checks
OpenClaw advisory guardian hook package: hooks/clawsec-advisory-guardian/
Setup scripts for hook and optional cron scheduling: scripts/
Guarded installer: scripts/guarded_skill_install.mjs
openclaw-audit-watchdog
soul-guardian
clawtributor (explicit opt-in)
npx clawhub@latest install clawsec-suite
set -euo pipefail
VERSION="${SKILL_VERSION:?Set SKILL_VERSION (e.g. 0.0.8)}"
INSTALL_ROOT="${INSTALL_ROOT:-$HOME/.openclaw/skills}"
DEST="$INSTALL_ROOT/clawsec-suite"
BASE="<https://github.com/prompt-security/clawsec/releases/download/clawsec-suite-v${VERSION}">
TEMP_DIR="$(mktemp -d)"
DOWNLOAD_DIR="$TEMP_DIR/downloads"
trap 'rm -rf "$TEMP_DIR"' EXIT
mkdir -p "$DOWNLOAD_DIR"
# 1) Download checksums manifest
curl -fsSL "$BASE/checksums.json" -o "$TEMP_DIR/checksums.json"
if ! jq -e '.skill and .version and .files' "$TEMP_DIR/checksums.json" >/dev/null 2>&1; then
echo "ERROR: Invalid checksums.json format" >&2
exit 1
fi
# 2) Download every file listed in checksums and verify immediately
DOWNLOAD_FAILED=0
for file in $(jq -r '.files | keys[]' "$TEMP_DIR/checksums.json"); do
FILE_URL="$(jq -r --arg f "$file" '.files[$f].url' "$TEMP_DIR/checksums.json")"
EXPECTED="$(jq -r --arg f "$file" '.files[$f].sha256' "$TEMP_DIR/checksums.json")"
if ! curl -fsSL "$FILE_URL" -o "$DOWNLOAD_DIR/$file"; then
echo "ERROR: Download failed for $file" >&2
DOWNLOAD_FAILED=1
continue
fi
if command -v shasum >/dev/null 2>&1; then
ACTUAL="$(shasum -a 256 "$DOWNLOAD_DIR/$file" | awk '{print $1}')"
else
ACTUAL="$(sha256sum "$DOWNLOAD_DIR/$file" | awk '{print $1}')"
fi
if [ "$EXPECTED" != "$ACTUAL" ]; then
echo "ERROR: Checksum mismatch for $file" >&2
DOWNLOAD_FAILED=1
else
echo "Verified: $file"
fi
done
if [ "$DOWNLOAD_FAILED" -eq 1 ]; then
echo "ERROR: One or more files failed verification" >&2
exit 1
fi
# 3) Install files using paths from checksums.json
while IFS= read -r file; do
[ -z "$file" ] && continue
REL_PATH="$(jq -r --arg f "$file" '.files[$f].path // $f' "$TEMP_DIR/checksums.json")"
SRC_PATH="$DOWNLOAD_DIR/$file"
DST_PATH="$DEST/$REL_PATH"
mkdir -p "$(dirname "$DST_PATH")"
cp "$SRC_PATH" "$DST_PATH"
done < <(jq -r '.files | keys[]' "$TEMP_DIR/checksums.json")
chmod 600 "$DEST/skill.json"
find "$DEST" -type f ! -name "skill.json" -exec chmod 644 {} \;
echo "Installed clawsec-suite v${VERSION} to: $DEST"
echo "Next step (OpenClaw): node \"$DEST/scripts/setup_advisory_hook.mjs\""
After installing the suite, enable the advisory guardian hook:
SUITE_DIR="${INSTALL_ROOT:-$HOME/.openclaw/skills}/clawsec-suite"
node "$SUITE_DIR/scripts/setup_advisory_hook.mjs"
Optional: create/update a periodic cron nudge (default every 6h) that triggers a main-session advisory scan:
SUITE_DIR="${INSTALL_ROOT:-$HOME/.openclaw/skills}/clawsec-suite"
node "$SUITE_DIR/scripts/setup_advisory_cron.mjs"
What this adds:
scan on agent:bootstrap and /new (command:new),
compare advisory affected entries against installed skills,
notify when new matches appear,
and ask for explicit user approval before any removal flow.
Restart the OpenClaw gateway after enabling the hook. Then run /new once to force an immediate scan in the next session context.
When the user asks to install a skill, treat that as the first request and run a guarded install check:
SUITE_DIR="${INSTALL_ROOT:-$HOME/.openclaw/skills}/clawsec-suite"
node "$SUITE_DIR/scripts/guarded_skill_install.mjs" --skill helper-plus --version 1.0.1
Behavior:
If no advisory match is found, install proceeds.
If advisory match is found, the script prints advisory context and exits with code 42.
Then require an explicit second confirmation from the user and rerun with --confirm-advisory:
node "$SUITE_DIR/scripts/guarded_skill_install.mjs" --skill helper-plus --version 1.0.1 --confirm-advisory
This enforces:
First confirmation: user asked to install.
Second confirmation: user explicitly approves install after seeing advisory details.
The embedded feed logic uses these defaults:
Remote feed URL: <https://raw.githubusercontent.com/prompt-security/clawsec/main/advisories/feed.json>
Local seed fallback: ~/.openclaw/skills/clawsec-suite/advisories/feed.json
State file: ~/.openclaw/clawsec-suite-feed-state.json
Hook rate-limit env (OpenClaw hook): CLAWSEC_HOOK_INTERVAL_SECONDS (default 300)
FEED_URL="${CLAWSEC_FEED_URL:-<https://raw.githubusercontent.com/prompt-security/clawsec/main/advisories/feed.json}">
STATE_FILE="${CLAWSEC_SUITE_STATE_FILE:-$HOME/.openclaw/clawsec-suite-feed-state.json}"
TMP="$(mktemp -d)"
trap 'rm -rf "$TMP"' EXIT
if ! curl -fsSLo "$TMP/feed.json" "$FEED_URL"; then
echo "ERROR: Failed to fetch advisory feed"
exit 1
fi
if ! jq -e '.version and (.advisories | type == "array")' "$TMP/feed.json" >/dev/null; then
echo "ERROR: Invalid advisory feed format"
exit 1
fi
mkdir -p "$(dirname "$STATE_FILE")"
if [ ! -f "$STATE_FILE" ]; then
echo '{"schema_version":"1.0","known_advisories":[],"last_feed_check":null,"last_feed_updated":null}' > "$STATE_FILE"
chmod 600 "$STATE_FILE"
fi
NEW_IDS_FILE="$TMP/new_ids.txt"
jq -r --argfile state "$STATE_FILE" '($state.known_advisories // []) as $known | [.advisories[]?.id | select(. != null and ($known | index(.) | not))] | .[]?' "$TMP/feed.json" > "$NEW_IDS_FILE"
if [ -s "$NEW_IDS_FILE" ]; then
echo "New advisories detected:"
while IFS= read -r id; do
[ -z "$id" ] && continue
jq -r --arg id "$id" '.advisories[] | select(.id == $id) | "- [\(.severity | ascii_upcase)] \(.id): \(.title)"' "$TMP/feed.json"
done < "$NEW_IDS_FILE"
else
echo "FEED_OK - no new advisories"
fi
Use the suite heartbeat script as the single periodic security check entrypoint:
skills/clawsec-suite/HEARTBEAT.mdIt handles:
suite update checks,
feed polling,
new-advisory detection,
affected-skill cross-referencing,
approval-gated response guidance for malicious/removal advisories,
and persistent state updates.
If an advisory indicates a malicious or removal-recommended skill and that skill is installed:
Notify the user immediately with advisory details and severity.
Recommend removing or disabling the affected skill.
Treat the original install request as first intent only.
Ask for explicit second confirmation before deletion/disable action (or before proceeding with risky install).
Only proceed after that second confirmation.
The suite hook and heartbeat guidance are intentionally non-destructive by default.
Install additional protections as needed:
npx clawhub@latest install openclaw-audit-watchdog
npx clawhub@latest install soul-guardian
# opt-in only:
npx clawhub@latest install clawtributor
Always verify checksums before installing files manually.
Keep advisory polling rate-limited (at least 5 minutes between checks).
Treat critical and high advisories affecting installed skills as immediate action items.
If you migrate off standalone clawsec-feed, keep one canonical state file to avoid duplicate notifications.