Security

hermes-attestation-guardian

Try it

Hermes-only runtime security attestation and drift detection skill for operator-managed Hermes infrastructure.

What it does

IMPORTANT SCOPE: This skill targets Hermes infrastructure only (CLI/Gateway/profile-managed deployments). This skill is not an OpenClaw runtime hook package.

The skill document

Hermes Attestation Guardian

IMPORTANT SCOPE:

  • This skill targets Hermes infrastructure only (CLI/Gateway/profile-managed deployments).
  • This skill is not an OpenClaw runtime hook package.

Vercel Skills Installation

Install with the Vercel Skills CLI for this harness:

npx skills add prompt-security/clawsec --skill hermes-attestation-guardian -a hermes-agent -y

Release Artifact Verification

For standalone installs, verify the signed release manifest before trusting SKILL.md, skill.json, or the archive. The skill.json file is the package metadata/SBOM source, and the release pipeline signs checksums.json with the ClawSec release key.

set -euo pipefail

SKILL_NAME="hermes-attestation-guardian"
VERSION="0.1.7"
REPO="prompt-security/clawsec"
TAG="${SKILL_NAME}-v${VERSION}"
BASE="https://github.com/${REPO}/releases/download/${TAG}"
ZIP_NAME="${SKILL_NAME}-v${VERSION}.zip"
TMP_DIR="$(mktemp -d)"
trap 'rm -rf "$TMP_DIR"' EXIT

RELEASE_PUBKEY_SHA256="711424e4535f84093fefb024cd1ca4ec87439e53907b305b79a631d5befba9c8"

curl -fsSL "$BASE/checksums.json" -o "$TMP_DIR/checksums.json"
curl -fsSL "$BASE/checksums.sig" -o "$TMP_DIR/checksums.sig"
curl -fsSL "$BASE/signing-public.pem" -o "$TMP_DIR/signing-public.pem"
curl -fsSL "$BASE/$ZIP_NAME" -o "$TMP_DIR/$ZIP_NAME"
curl -fsSL "$BASE/SKILL.md" -o "$TMP_DIR/SKILL.md"
curl -fsSL "$BASE/skill.json" -o "$TMP_DIR/skill.json"

ACTUAL_PUBKEY_SHA256="$(openssl pkey -pubin -in "$TMP_DIR/signing-public.pem" -outform DER | shasum -a 256 | awk '{print $1}')"
if [ "$ACTUAL_PUBKEY_SHA256" != "$RELEASE_PUBKEY_SHA256" ]; then
  echo "ERROR: signing-public.pem fingerprint mismatch" >&2
  exit 1
fi

openssl base64 -d -A -in "$TMP_DIR/checksums.sig" -out "$TMP_DIR/checksums.sig.bin"
openssl pkeyutl -verify -rawin -pubin \
  -inkey "$TMP_DIR/signing-public.pem" \
  -sigfile "$TMP_DIR/checksums.sig.bin" \
  -in "$TMP_DIR/checksums.json" >/dev/null

hash_file() {
  if command -v shasum >/dev/null 2>&1; then
    shasum -a 256 "$1" | awk '{print $1}'
  else
    sha256sum "$1" | awk '{print $1}'
  fi
}

verify_manifest_file() {
  asset="$1"
  path="$2"
  expected="$(jq -r --arg asset "$asset" '.files[$asset].sha256 // empty' "$TMP_DIR/checksums.json")"
  if [ -z "$expected" ]; then
    echo "ERROR: checksums.json missing $asset" >&2
    exit 1
  fi
  actual="$(hash_file "$path")"
  if [ "$actual" != "$expected" ]; then
    echo "ERROR: checksum mismatch for $asset" >&2
    exit 1
  fi
}

expected_archive="$(jq -r '.archive.sha256 // empty' "$TMP_DIR/checksums.json")"
if [ -z "$expected_archive" ]; then
  echo "ERROR: checksums.json missing archive.sha256" >&2
  exit 1
fi
actual_archive="$(hash_file "$TMP_DIR/$ZIP_NAME")"
if [ "$actual_archive" != "$expected_archive" ]; then
  echo "ERROR: archive checksum mismatch" >&2
  exit 1
fi

verify_manifest_file "SKILL.md" "$TMP_DIR/SKILL.md"
verify_manifest_file "skill.json" "$TMP_DIR/skill.json"

echo "Signed release manifest, archive, SKILL.md, and skill.json verified."

Only install or extract the archive after this verification succeeds.

Goal

Generate deterministic Hermes posture attestations, verify them with fail-closed integrity checks, and compare baseline drift using stable severity mapping.

Hermes guard trust policy note

When installing from community sources, configure Hermes guard to use signature-aware trust (trusted signer fingerprint allowlist) rather than source-name-only trust. Unknown signer fingerprints should stay on community policy, and invalid signatures must remain blocked.

Commands

# Generate attestation (default output: ~/.hermes/security/attestations/current.json)
node scripts/generate_attestation.mjs

# Generate with explicit policy + deterministic timestamp
node scripts/generate_attestation.mjs \
  --policy ~/.hermes/security/attestation-policy.json \
  --generated-at 2026-04-15T18:00:00.000Z \
  --write-sha256

# Verify schema + canonical digest
node scripts/verify_attestation.mjs --input ~/.hermes/security/attestations/current.json

# Verify with baseline diff (baseline must be authenticated)
node scripts/verify_attestation.mjs \
  --input ~/.hermes/security/attestations/current.json \
  --baseline ~/.hermes/security/attestations/baseline.json \
  --baseline-expected-sha256  \
  --fail-on-severity high

# Optional detached signature verification
node scripts/verify_attestation.mjs \
  --input ~/.hermes/security/attestations/current.json \
  --signature ~/.hermes/security/attestations/current.json.sig \
  --public-key ~/.hermes/security/keys/attestation-public.pem

# Refresh advisory feed verification state (fail-closed by default)
node scripts/refresh_advisory_feed.mjs

# Check advisory feed verification + feed summary
node scripts/check_advisories.mjs

# Guarded advisory-aware skill verification gate (returns 42 on advisory match without explicit confirm)
node scripts/guarded_skill_verify.mjs --skill some-skill --version 1.2.3

# Explicit operator acknowledgement path for advisory matches
node scripts/guarded_skill_verify.mjs --skill some-skill --version 1.2.3 --confirm-advisory

# Optional temporary unsigned bypass (dangerous; emergency-only)
HERMES_ADVISORY_ALLOW_UNSIGNED_FEED=1 node scripts/refresh_advisory_feed.mjs --allow-unsigned

# Preview scheduler config without mutating user schedule state
node scripts/setup_attestation_cron.mjs --every 6h --print-only

# Apply managed scheduler block
node scripts/setup_attestation_cron.mjs --every 6h --apply

# Preview advisory check scheduler config (guarded flow, print-only default)
node scripts/setup_advisory_check_cron.mjs --every 6h --skill some-skill --print-only

# Apply advisory check scheduler block (uses guarded_skill_verify flow)
node scripts/setup_advisory_check_cron.mjs --every 6h --skill some-skill --version 1.2.3 --apply

# Emergency-only: unsigned bypass for scheduled advisory checks (do not keep enabled)
node scripts/setup_advisory_check_cron.mjs --every 6h --skill some-skill --allow-unsigned --apply

WARNING: --allow-unsigned in scheduled commands is incident-response only. Remove it immediately after recovery and restore signed advisory verification.

Attestation payload (implemented)

The generator emits:

  • schema_version, platform, generated_at
  • generator metadata (skill + node version)
  • host metadata (hostname/platform/arch)
  • posture.runtime (gateway enabled flags + risky toggles)
  • posture.feed_verification status (verified|unverified|unknown) sourced from $HERMES_HOME/security/advisories/feed-verification-state.json
  • posture.integrity watched_files and trust_anchors (existence + sha256)
  • digests.canonical_sha256 over a stable canonical JSON representation

Fail-closed behavior

Verifier exits non-zero when:

  • schema validation fails
  • canonical digest algorithm is unsupported or digest binding mismatches
  • expected file sha256 mismatches (if configured)
  • detached signature verification fails (if configured)
  • baseline is provided without authenticated trust binding (--baseline-expected-sha256 and/or baseline signature + public key)
  • baseline authenticity or baseline schema/digest validation fails
  • baseline diff highest severity is at/above --fail-on-severity (default: critical)

Severity messages are emitted as INFO / WARNING / CRITICAL style lines.

Side effects

  • generate_attestation.mjs writes one JSON file (and optional .sha256) under $HERMES_HOME/security/attestations.
  • verify_attestation.mjs is read-only.
  • refresh_advisory_feed.mjs writes verified feed cache + verification state under $HERMES_HOME/security/advisories.
  • check_advisories.mjs is read-only.
  • guarded_skill_verify.mjs re-runs feed refresh/verification (same advisory cache + state side effects) and then performs advisory-aware gate checks.
  • setup_attestation_cron.mjs is read-only unless --apply is provided.
  • setup_attestation_cron.mjs --apply rewrites only the current user managed schedule block delimited by:
    • # >>> hermes-attestation-guardian >>>
    • # <<< hermes-attestation-guardian <<<
  • setup_advisory_check_cron.mjs is read-only unless --apply is provided.
  • setup_advisory_check_cron.mjs --apply rewrites only the current user advisory-check managed schedule block delimited by:
    • # >>> hermes-attestation-guardian-advisory-check >>>
    • # <<< hermes-attestation-guardian-advisory-check <<<
    • generated command path uses guarded_skill_verify.mjs (advisory-aware gate), not raw check_advisories.mjs

Advisory feed override knobs

The default signed advisory feed is consolidated: it can contain NVD CVEs, approved community advisories, and provisional GHSA-without-CVE records. Hermes matching still gates on affected package names and supported version ranges.

  • Source selection: HERMES_ADVISORY_FEED_SOURCE=auto|remote|local
  • Remote artifacts: HERMES_ADVISORY_FEED_URL, HERMES_ADVISORY_FEED_SIG_URL, HERMES_ADVISORY_FEED_CHECKSUMS_URL, HERMES_ADVISORY_FEED_CHECKSUMS_SIG_URL
  • Local artifacts: HERMES_LOCAL_ADVISORY_FEED, HERMES_LOCAL_ADVISORY_FEED_SIG, HERMES_LOCAL_ADVISORY_FEED_CHECKSUMS, HERMES_LOCAL_ADVISORY_FEED_CHECKSUMS_SIG
  • Pinned key override: HERMES_ADVISORY_FEED_PUBLIC_KEY (default is built-in pinned key)
  • Optional checksum toggle: HERMES_ADVISORY_VERIFY_CHECKSUM_MANIFEST (default: enabled)
  • UNSAFE emergency bypass only: HERMES_ADVISORY_ALLOW_UNSIGNED_FEED=1

Notes

  • Hermes scan + test context is .mjs-based by design:
    • runtime scripts: scripts/*.mjs
    • shared libraries: lib/*.mjs
    • regression tests: test/*.test.mjs
  • Keep .mjs paths/extensions stable so scanner scope, SBOM wiring, and test harness references stay valid.
  • Default output root is ~/.hermes/security/attestations/.
  • No destructive remediation actions (delete/restore/quarantine) are implemented.
  • Advisory feed remote URL allowlisting is not implemented in v0.0.2; operators must explicitly trust configured feed/checksum endpoints.
  • Guarded advisory version matching supports >=, <=, >, <, =, ^, ~, wildcard *, AND comparator sets separated by spaces or commas, and SemVer prerelease precedence. OR and hyphen ranges remain unsupported and fail closed.
  • Valid CPE 2.3 entries from the consolidated NVD feed are accepted as non-package metadata and ignored by package-name matching. Malformed CPE entries still fail closed.
  • Operator policy file is optional JSON with:
    • watch_files: list of file paths
    • trust_anchor_files: list of file paths

Related skills

Stores durable facts in a categorized, plain-markdown vault on disk, alongside your agent's built-in memory.

by Iván559 installs18 stars

Find why your productivity system keeps failing, then apply the smallest fix — capacity math, bottleneck routing, durable local notes.

by Iván855 installs69 stars

Join a video meeting as an AI bot with voice, avatar, and screenshare across four operating modes.

by johnpatternai22 installs8 stars

Generate and edit Draw.io, Mermaid, and Excalidraw diagrams from natural language using a structured JSON spec.

by nssa.io1.0k installs47 stars

Fetch raw ad creative, app, ranking, and revenue data from AdMapix as structured JSON.

by fly0pants4.3k installs296 stars

More from davida-ps

Browse all skills

Advisory-feed monitoring, signed-trust verification, and approval-gated response for OpenClaw skills, bundled in one suite.

by davida-ps359 installs8 stars

Subscribe AI agents to a community-curated security advisory feed for AI-agent threats.

by davida-ps103 installs1 stars

Automated daily security audits for OpenClaw agents with DM delivery and optional email reporting. Runs deep audits, creates or updates a recurring cron job,...

by davida-ps124 installs3 stars

Drift detection + baseline integrity guard for agent workspace files with automatic alerting support

by davida-ps152 installs1 stars

Community incident reporting for AI agents. Contribute to collective security by reporting threats.

by davida-ps92 installs1 stars

Use when checking for security vulnerabilities in NanoClaw skills, before installing new skills, or when asked about security advisories affecting the bot

by davida-ps48 installs