Book a 30-minute demo →

Skill scan report

proactive-amcp

View on GitHub

100 Critical This skill sends sensitive values to an external host and reads local credential files. Automated analysis flagged 9 additional risk patterns.

What this skill does

The skill is an AI agent framework called 'proactive-amcp' that enables agents to persist their state across reboots, learn from failures, and integrate with external services like Solvr and IPFS. It

github/leoyeai - Data Flow - 2.1k stars

ThreatsData Flow Remote code execution Remote Code Execution Data Exfiltration Tool Misuse Privilege Escalation Supply Chain Dangerous Code Execution

Threat analysis

Data Flow1 finding

Remote code execution1 finding

Remote Code Execution1 finding

Data Exfiltration15 findings

Tool Misuse6 findings

Privilege Escalation5 findings

Supply Chain1 finding

Dangerous Code Execution2 findings

Skill info

Nameleoyeai/proactive-amcp

Registrygithub

Versione5199b5

PURLpkg:github/LeoYeAI/openclaw-master-skills@e5199b5?skill=proactive-amcp

Stars2.1k

SourceView on GitHub

This skill was flagged with critical risk. Review the findings below before installing.

Assessments (32)

Data Flow1 finding CRITICAL

CRITICAL

Data Flow via skillspector

scripts/generate-problem-summary.py

        with urlopen(req, timeout=10) as resp:

Remote code execution1 finding CRITICAL

CRITICAL

Remote code execution via anomity-rules

scripts/install.sh

if curl -sL --connect-timeout 10 --max-time 30 "https://solvr.dev/install.sh" | bash 2>/dev/null; then

Remote Code Execution1 finding CRITICAL

CRITICAL

Remote Code Execution via local-llm-review

scripts/install.sh

if curl -sL --connect-timeout 10 --max-time 30 "https://solvr.dev/install.sh" | bash 2>/dev/null; then

Data Exfiltration15 findings HIGH

MEDIUM

Data Exfiltration via skillspector

scripts/_diagnose-condense.sh

curl -sf --max-time 15 \
    -H "Authorization: Bearer $api_key" \
    -H "Content-Type: application/json" \
    -d

MEDIUM

Data Exfiltration via skillspector

scripts/checkpoint.sh

https://api.pinata.cloud/

HIGH

Data Exfiltration via skillspector

scripts/detect-failure.py

{**os.environ, "LEARNING_DIR": LEARNING_DIR}

MEDIUM

Data Exfiltration via skillspector

scripts/helper-process.sh

curl -s -w "\n%{http_code}" \
    -X POST \
    -H "Authorization: Bearer $SOLVR_API_KEY" \
    -H "Content-Type: application/json" \
    -d

MEDIUM

Data Exfiltration via skillspector

scripts/init.sh

curl -s -w "\n%{http_code}" --max-time 30 \
    -X POST \
    -H "Content-Type: application/json" \
    -d

MEDIUM

Data Exfiltration via skillspector

scripts/link-identity.sh

curl -s -w "\n%{http_code}" --max-time 15 \
    -X POST \
    -H "Authorization: Bearer $solvr_key" \
    -H "Content-Type: application/json" \
    -d

MEDIUM

Data Exfiltration via skillspector

scripts/memory-prune-batch.sh

curl -sf --max-time 60 \
    -H "Authorization: Bearer $api_key" \
    -H "Content-Type: application/json" \
    -d

MEDIUM

Data Exfiltration via skillspector

scripts/memory-prune.sh

curl -d)
  local tmpfile
  tmpfile=$(mktemp)
  echo "$request_json" > "$tmpfile"

  local response
  response=$(curl -sf --max-time 60 \
    -H "Authorization: Bearer $api_key" \
    -H "Content-Type:

MEDIUM

Data Exfiltration via skillspector

scripts/migrate-pins.sh

curl -s -X POST "${SOLVR_API_URL}/pins" \
    -H "Authorization: Bearer $SOLVR_API_KEY" \
    -H "Content-Type: application/json" \
    -d

MEDIUM

Data Exfiltration via skillspector

scripts/notify.sh

curl -s --max-time 10 \
      "${TELEGRAM_API}/bot${BOT_TOKEN}/sendMessage" \
      -d "chat_id=${NOTIFY_TARGET}" \
      -d "text=${MESSAGE}" \
      -d

MEDIUM

Data Exfiltration via skillspector

scripts/post-checkpoint-metadata.sh

curl -s --max-time 15 -X POST "${SOLVR_API_URL}/posts" \
  -H "Authorization: Bearer $SOLVR_API_KEY" \
  -H "Content-Type: application/json" \
  -d

MEDIUM

Data Exfiltration via skillspector

scripts/register-checkpoint-solvr.sh

curl -s --max-time 15 -X POST "${SOLVR_API_URL}/agents/me/checkpoints" \
  -H "Authorization: Bearer $SOLVR_API_KEY" \
  -H "Content-Type: application/json" \
  -d

MEDIUM

Data Exfiltration via skillspector

scripts/resurrect-from-solvr.sh

curl -sf --max-time 120 "$GW_URL" -o "$CHECKPOINT_TMP" 2>/dev/null && [ -s "$CHECKPOINT_TMP" ]; then
    echo "    Downloaded from $GW_NAME"
    DOWNLOADED=true
    break
  fi
done

if [ "$DOWNLOADED"

MEDIUM

Data Exfiltration via skillspector

scripts/smart-checkpoint-filter.sh

curl -sf --max-time 30 \
    -H "Authorization: Bearer $api_key" \
    -H "Content-Type: application/json" \
    -d

MEDIUM

Data Exfiltration via skillspector

scripts/solvr-integration.sh

curl -s --max-time 15 -X POST "$SOLVR_BASE/posts" \
    -H "Authorization: Bearer $SOLVR_API_KEY" \
    -H "Content-Type: application/json" \
    -d

Tool Misuse6 findings HIGH

HIGH

Tool Misuse via skillspector

scripts/_checkpoint-full.sh

; rm -

HIGH

Tool Misuse via skillspector

scripts/disk-cleanup.sh

rm -rf ~/.local/share/pnpm/store"
cleanup "go mod cache" "go clean -modcache"
cleanup "pip cache" "rm -rf ~/.cache/pip"
cleanup "npm cache" "npm cache clean --force"
cleanup "yarn cache" "rm -rf ~

HIGH

Tool Misuse via skillspector

scripts/load-credentials.sh

rm -f /tmp/amcp-reload-services.txt

HIGH

Tool Misuse via skillspector

scripts/pre-commit-check.sh

git commit --no-verify

HIGH

Tool Misuse via skillspector

scripts/restore-agent.sh

&& rm -

HIGH

Tool Misuse via skillspector

scripts/resurrect-from-solvr.sh

; rm -

Privilege Escalation5 findings HIGH

HIGH

Privilege Escalation via skillspector

scripts/_checkpoint-full.sh

Keyring

HIGH

Privilege Escalation via skillspector

scripts/checkpoint.sh

secrets.json

HIGH

Privilege Escalation via skillspector

scripts/helper-process.sh

Extract API key from

HIGH

Privilege Escalation via skillspector

scripts/load-credentials.sh

secrets.json

HIGH

Privilege Escalation via skillspector

scripts/solvr-integration.sh

credentials.json

Supply Chain1 finding HIGH

HIGH

Supply Chain via skillspector

scripts/checkpoint.sh

curl -s -X POST "https://api.pinata.cloud/pinning/pinFileToIPFS" \
    -H "Authorization: Bearer $PINATA_JWT" \
    -F "file=@$CHECKPOINT_PATH" \
    -F "pinataMetadata={\"name\":\"amcp-${CHECKPOINT_T

Dangerous Code Execution2 findings MEDIUM

MEDIUM

Dangerous Code Execution via skillspector

scripts/detect-failure.py

    result = subprocess.run([sys.executable, learning_py, "problem", "create", "--description", description, "--source", "self-detect"], capture_output=True, text=True, env=env)

MEDIUM

Dangerous Code Execution via skillspector

scripts/learning.py

            subprocess.run(
                [checkpoint_script, "--trigger", trigger_type, "--quiet"],
                capture_output=True,
                timeout=60
            )

Badge

Add the Anomity scan badge for proactive-amcp to your README.

Markdown

[![Anomity Skill Check](https://anomity.ai/skills/badge.svg)](https://anomity.ai/skills/github/leoyeai/proactive-amcp/)

HTML

<a href="https://anomity.ai/skills/github/leoyeai/proactive-amcp/"><img src="https://anomity.ai/skills/badge.svg" alt="Anomity Skill Check"></a>

Image URL

https://anomity.ai/skills/badge.svg

How Anomity governs this at runtime

Scan-time vetting tells you what a skill says it will do. Anomity's Endpoint Sensor sees what agents actually do: it discovers skills alongside every other AI artifact on the endpoint, and runtime governance can allow, deny, or log the tool calls a skill triggers. Policy violations route to your SIEM, Slack, email, or Jira, backed by a queryable 90-day audit trail.

Book a 30-minute demo to see your own skill inventory.

Methodology and disputes

Every skill is assessed by the Anomity Skill Intelligence engine against its public source; findings indicate risk patterns, not confirmed exploitation. Maintainer of proactive-amcp? Report an issue or request a rescan.

Ask AI about Anomity