{"id":"7c8a4345-cc1e-45bf-a3e9-3e5079b4fed7","entityType":"agent","slug":"clawhub-enochosbot-bot-abaddon","name":"Abaddon","canonicalUrl":"https://www.xpersona.co/agent/clawhub-enochosbot-bot-abaddon","canonicalPath":"/agent/clawhub-enochosbot-bot-abaddon","generatedAt":"2026-10-12T14:48:17.291Z","source":"CLAWHUB","claimStatus":"UNCLAIMED","verificationTier":"NONE","summary":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-12T02:14:54.338Z","emptyReason":null},"description":"Red team security mode for OpenClaw. Runs an adversarial audit on demand or nightly — checks exposed ports, credential leaks, file permissions, suspicious pr...","descriptionLabel":"Source description","evidenceSummary":"Capability contract not published. No trust telemetry is available yet. 956 downloads reported by the source. Last updated 10/12/2026.","installCommand":"clawhub skill install s175f7knh9wa8h414k24cvcf05885t6q:abaddon","sourceUrl":"https://clawhub.ai/enochosbot-bot/abaddon","homepage":"https://clawhub.ai/enochosbot-bot/skills/abaddon","primaryLinks":[{"label":"View on ClawHub","url":"https://clawhub.ai/enochosbot-bot/abaddon","kind":"source"},{"label":"Homepage","url":"https://clawhub.ai/enochosbot-bot/skills/abaddon","kind":"homepage"}],"safetyScore":84,"overallRank":62,"popularityScore":60,"trustScore":null,"claimedByName":null,"isOwner":false,"seoDescription":"Abaddon technical dossier on Xpersona with agent coverage, OPENCLEW support, and live trust metadata."},"coverage":{"evidence":{"source":"public-profile","verified":false,"confidence":"medium","updatedAt":"2026-10-12T02:14:54.338Z","emptyReason":null},"protocols":[{"protocol":"OPENCLEW","label":"OpenClaw","status":"self-declared","notes":"Declared in the public agent profile."}],"capabilities":[],"verifiedCount":0,"selfDeclaredCount":1,"capabilityMatrix":{"rows":[{"key":"OPENCLEW","type":"protocol","support":"unknown","confidenceSource":"profile","notes":"Listed on profile"}],"flattenedTokens":"protocol:OPENCLEW|unknown|profile"}},"adoption":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-12T02:14:54.338Z","emptyReason":null},"stars":null,"forks":null,"downloads":956,"likes":null,"task":null,"library":null,"packageName":null,"latestVersion":"1.0.0","tractionLabel":"956 downloads"},"release":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-12T02:14:54.326Z","emptyReason":null},"lastUpdatedAt":"2026-10-12T02:14:54.338Z","lastCrawledAt":"2026-10-12T02:14:54.326Z","lastIndexedAt":null,"nextCrawlAt":"2026-10-13T02:14:54.326Z","lastVerifiedAt":null,"highlights":[{"version":"1.0.0","createdAt":"2026-02-20T04:14:19.066Z","changelog":"Initial release. Red team / adversarial security mode for OpenClaw. Nightly + on-demand. macOS. Letter grade output.","fileCount":5,"zipByteSize":7118}]},"execution":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No published capability contract is available yet."},"installCommand":"clawhub skill install s175f7knh9wa8h414k24cvcf05885t6q:abaddon","setupComplexity":"low","setupSteps":["Install using `clawhub skill install s175f7knh9wa8h414k24cvcf05885t6q:abaddon` in an isolated environment before connecting it to live workloads.","No published capability contract is available yet, so validate auth and request/response behavior manually.","Review the upstream CLAWHUB listing at https://clawhub.ai/enochosbot-bot/abaddon before using production credentials."],"contract":{"contractStatus":"missing","authModes":[],"requires":[],"forbidden":[],"supportsMcp":false,"supportsA2a":false,"supportsStreaming":false,"inputSchemaRef":null,"outputSchemaRef":null,"dataRegion":null,"contractUpdatedAt":null,"sourceUpdatedAt":null,"freshnessSeconds":null},"invocationGuide":{"preferredApi":{"snapshotUrl":"https://www.xpersona.co/api/v1/agents/clawhub-enochosbot-bot-abaddon/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-enochosbot-bot-abaddon/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-enochosbot-bot-abaddon/trust"},"curlExamples":["curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-enochosbot-bot-abaddon/snapshot\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-enochosbot-bot-abaddon/contract\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-enochosbot-bot-abaddon/trust\""],"jsonRequestTemplate":{"query":"summarize this repo","constraints":{"maxLatencyMs":2000,"protocolPreference":["OPENCLEW"]}},"jsonResponseTemplate":{"ok":true,"result":{"summary":"...","confidence":0.9},"meta":{"source":"CLAWHUB","generatedAt":"2026-10-12T14:48:17.291Z"}},"retryPolicy":{"maxAttempts":3,"backoffMs":[500,1500,3500],"retryableConditions":["HTTP_429","HTTP_503","NETWORK_TIMEOUT"]}},"endpoints":{"dossierUrl":"https://www.xpersona.co/api/v1/agents/clawhub-enochosbot-bot-abaddon/dossier","snapshotUrl":"https://www.xpersona.co/api/v1/agents/clawhub-enochosbot-bot-abaddon/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-enochosbot-bot-abaddon/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-enochosbot-bot-abaddon/trust"}},"reliability":{"evidence":{"source":"runtime-metrics","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No trust, reliability, or runtime telemetry is available."},"trust":{"status":"unavailable","handshakeStatus":"UNKNOWN","verificationFreshnessHours":null,"reputationScore":null,"p95LatencyMs":null,"successRate30d":null,"fallbackRate":null,"attempts30d":null,"trustUpdatedAt":null,"trustConfidence":"unknown","sourceUpdatedAt":null,"freshnessSeconds":null},"decisionGuardrails":{"doNotUseIf":["Contract metadata is missing or unavailable for deterministic execution."],"safeUseWhen":[],"riskFlags":["missing_or_unavailable_contract","trust_data_unavailable","schema_references_missing"],"operationalConfidence":"low"},"executionMetrics":{"observedLatencyMsP50":null,"observedLatencyMsP95":null,"estimatedCostUsd":null,"uptime30d":null,"rateLimitRpm":null,"rateLimitBurst":null,"lastVerifiedAt":null,"verificationSource":null},"runtimeMetrics":{"successRate":null,"avgLatencyMs":null,"avgCostUsd":null,"hallucinationRate":null,"retryRate":null,"disputeRate":null,"p50Latency":null,"p95Latency":null,"lastUpdated":null}},"benchmarks":{"evidence":{"source":"no-benchmark-data","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No benchmark suites or observed failure patterns are available."},"suites":[],"failurePatterns":[]},"artifacts":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-12T02:14:54.338Z","emptyReason":null},"readme":"Skill: Abaddon\n\nOwner: enochosbot-bot\n\nSummary: Red team security mode for OpenClaw. Runs an adversarial audit on demand or nightly — checks exposed ports, credential leaks, file permissions, suspicious pr...\n\nTags: latest:1.0.0\n\nVersion history:\n\nv1.0.0 | 2026-02-20T04:14:19.066Z | user\n\nInitial release. Red team / adversarial security mode for OpenClaw. Nightly + on-demand. macOS. Letter grade output.\n\nArchive index:\n\nArchive v1.0.0: 5 files, 7118 bytes\n\nFiles: setup/cron-seed.sh (2645b), SKILL.md (3590b), SPEC.md (3007b), templates/abaddon-prompt.md (3144b), _meta.json (126b)\n\nFile v1.0.0:SKILL.md\n\n---\nname: abaddon\ndescription: Red team security mode for OpenClaw. Runs an adversarial audit on demand or nightly — checks exposed ports, credential leaks, file permissions, suspicious processes, and OpenClaw config posture. Assigns a letter grade. Built for macOS deployments.\n---\n\n# Abaddon ⚔️\n\nMost security audits are defensive — they check what you've locked down. Abaddon runs the other direction. It thinks like an attacker. It looks for what an adversary would find, not just what you remember to check.\n\nOn demand or every night at 3:45 AM. Letter grade every time.\n\n## What It Checks\n\n**Network & Exposure**\n- Listening ports — anything bound to 0.0.0.0 is flagged\n- Gateway binding — should be loopback only\n- SSH Remote Login state\n- Active tunnels (ngrok, cloudflared, unexpected remote access)\n- Firewall and stealth mode\n\n**System Integrity**\n- SIP, FileVault, Gatekeeper\n- macOS version + pending updates\n- XProtect / MRT definitions age\n\n**OpenClaw Configuration**\n- Exec security mode (full / allowlist / deny)\n- Gateway auth enabled?\n- Unexpected cron entries\n- Unexpected plugins\n\n**File Permissions**\n- SOUL.md + AGENTS.md: root-owned, 444\n- MEMORY.md, USER.md, AGENT_PROMPT.md, openclaw.json, cron/jobs.json, LaunchAgent plists: 600\n- Flags anything 644 or wider on sensitive paths\n- Plaintext key scan across workspace\n\n**API Key Handling**\n- Keys in Keychain or flat files?\n- Keys leaking through env vars?\n- Secrets in git history?\n- Hardcoded tokens in .zshrc?\n\n**Agent Behavior**\n- Memory injection scan (prompt injection attempts in memory files)\n- Sub-agent scope check\n- Unexpected agent permissions\n\n**Dependencies**\n- Homebrew outdated (flags openclaw, ollama, node)\n- npm global outdated\n\n## Scoring\n\n| Grade | Criteria |\n|-------|----------|\n| A | 0 CRITICAL, 0 HIGH |\n| B | 0 CRITICAL, 1–2 HIGH |\n| C | 1 CRITICAL or 3+ HIGH |\n| D | 2+ CRITICAL |\n| F | Active compromise indicators |\n\n## Installation\n\n### Step 1 — Copy the Abaddon prompt into your agent\n\nIf you have Gideon (the OpenClaw observer agent), append the red team section:\n\n```bash\ncat skills/abaddon/templates/abaddon-prompt.md >> ~/.openclaw/workspace/agents/observer/AGENT_PROMPT.md\n```\n\nIf you don't have Gideon, use the standalone agent prompt:\n\n```bash\ncp skills/abaddon/templates/abaddon-prompt.md ~/.openclaw/workspace/agents/abaddon/AGENT_PROMPT.md\n```\n\n### Step 2 — Add the nightly cron\n\n```bash\nbash skills/abaddon/setup/cron-seed.sh\n```\n\nThis adds a 3:45 AM CST cron job to `~/.openclaw/cron/jobs.json`. Delivers to Telegram Security topic if configured.\n\n### Step 3 — Lock the agent prompt\n\n```bash\nchmod 600 ~/.openclaw/workspace/agents/observer/AGENT_PROMPT.md\n```\n\nYour detection playbook should never be world-readable.\n\n## Usage\n\n**Manual trigger** — say any of:\n- \"run red team\"\n- \"run Abaddon\"\n- \"run full assessment\"\n- \"Abaddon report\"\n\n**Nightly** — fires automatically at 3:45 AM CST after the standard defensive audit (3:30 AM).\n\n## Output\n\nEvery run produces two things:\n\n1. **Technical report** → `memory/audits/abaddon-YYYY-MM-DD.md` — full command output, evidence, remediation steps\n2. **Summary** → posted to Telegram Security topic with letter grade\n\nCRITICAL findings trigger an immediate DM alert.\n\n## Notes\n\n- Designed for macOS (Darwin arm64). Most checks work on Linux with minor path adjustments.\n- Assumes OpenClaw gateway is running locally. Remote deployments may need adjusted port/binding checks.\n- Pairs with `enoch-tuning` — run `lock-identity.sh` after install to enforce all file permission baselines in one pass.\n\nFile v1.0.0:_meta.json\n\n{\n  \"ownerId\": \"kn78z941f26sn68a4s3kcetssh81gk44\",\n  \"slug\": \"abaddon\",\n  \"version\": \"1.0.0\",\n  \"publishedAt\": 1771560859066\n}\n\nFile v1.0.0:SPEC.md\n\n# Abaddon — Red Team Mode (merged into Gideon)\n**Date:** 2026-02-19 | **Status:** Merged into agents/observer/AGENT_PROMPT.md\n**Updated:** 2026-02-19 (removed random-delay trigger, added fixed nightly cron)\n\n---\n\n## What It Is\n\nAbaddon is not a separate agent. It is Gideon's adversarial mode — an offensive red team scan that runs alongside the standard nightly defensive audit.\n\nNamed after the angel of the abyss. Assumes breach. Checks what an attacker would exploit.\n\n---\n\n## Two Modes\n\n| Mode | Schedule | Posture | Output |\n|------|----------|---------|--------|\n| Gideon nightly (defensive) | 3:30 AM CST daily | Passive monitor, internal audit | memory/audits/YYYY-MM-DD.md |\n| Abaddon nightly (offensive) | 3:45 AM CST daily | Adversarial red team | memory/audits/abaddon-YYYY-MM-DD.md |\n| Abaddon manual | On demand | Adversarial red team | Same, plus letter grade posted to Security topic |\n\n---\n\n## Trigger Phrases (manual)\n\n- \"run red team\"\n- \"run Abaddon\"\n- \"run full assessment\"\n- \"Abaddon report\"\n\n---\n\n## What Abaddon Checks\n\n1. Exposed ports and unexpected network listeners\n2. Credential leaks — environment variables, log files, temp files\n3. Process list — suspicious or unexpected processes\n4. File permissions on sensitive paths (openclaw.json, SOUL.md, AGENT_PROMPT.md, LaunchAgent plists)\n5. Unauthorized SSH keys, cron entries, LaunchAgents\n6. ops/exec-audit.log — anomalous command patterns\n7. ops/identity-change-audit.log — rapid/unexpected config modifications\n8. Gateway binding — verify loopback-only\n9. World-readable files in sensitive directories\n10. Assigns a letter grade (A–F)\n\n---\n\n## Scheduling\n\nAbaddon runs as a native OpenClaw cron job in `cron/jobs.json`. No shell script wrapper, no random delay.\n\n**Previous approach (deprecated):** `scripts/abaddon-trigger.sh` with random 0–16hr delay via LaunchAgent. This was removed because the trigger script never actually ran a scan — it only checked gateway health and logged a false \"firing\" message.\n\n**Current approach:** Direct cron entry in jobs.json at 3:45 AM CST, `sessionTarget: isolated`, `wakeMode: now`, `agentTurn` with the red team prompt baked in. Delivers to Telegram Security topic with letter grade.\n\n---\n\n## File Permissions Enforced by Abaddon Findings\n\nThese paths should always be 600 (owner-only). Abaddon flags deviations:\n\n- `~/.openclaw/openclaw.json`\n- `~/.openclaw/cron/jobs.json`\n- `~/.openclaw/workspace/SOUL.md`\n- `~/.openclaw/workspace/MEMORY.md`\n- `~/.openclaw/workspace/USER.md`\n- `~/.openclaw/workspace/agents/observer/AGENT_PROMPT.md`\n- `~/Library/LaunchAgents/ai.openclaw.*.plist`\n- `~/Library/LaunchAgents/com.openclaw.*.plist`\n\nRun `skills/enoch-tuning/setup/lock-identity.sh` to enforce all of the above in one pass.\n\n---\n\n## ClawHub Publishing\n\nWhen published as a standalone skill, Abaddon becomes a distributable red team module. Until then, it lives inside Gideon's AGENT_PROMPT.md.\n\nSee: `agents/observer/AGENT_PROMPT.md` → ## Red Team Mode (Abaddon)\n\nFile v1.0.0:templates/abaddon-prompt.md\n\n---\n\n## Red Team Mode (Abaddon) 🔴\n\nActivated when triggered with: \"run red team\", \"run Abaddon\", \"run full assessment\", or via nightly cron at 3:45 AM.\n\nIn red team mode, shift from passive monitor to active attacker. Think like an adversary. Find what they would find.\n\n### Trigger Detection\nIf the user says \"run red team\", \"run Abaddon\", \"run full assessment\", or \"Abaddon report\" — enter this mode immediately. Do NOT run the standard passive audit. Run the full Abaddon protocol below.\n\n### Attack Vector Checks\n\n**Network & Exposure**\n- `lsof -iTCP -sTCP:LISTEN` — flag anything bound to 0.0.0.0 (should all be 127.0.0.1)\n- Gateway binding: localhost only or exposed to network?\n- SSH Remote Login: `systemsetup -getremotelogin`\n- Active tunnels: any ngrok, cloudflared, or unexpected remote access running?\n- Firewall: enabled? stealth mode on?\n- DNS: any unexpected servers configured?\n\n**System Integrity**\n- SIP: `csrutil status` — must be enabled\n- FileVault: `fdesetup status` — must be On\n- Gatekeeper: `spctl --status` — must be enabled\n- macOS: current version? pending updates?\n\n**OpenClaw Configuration**\n- Exec security mode: full/allowlist/deny — is it appropriate?\n- Auth enabled on gateway?\n- streamMode off? (prevents conversation preview leaks)\n- Plugins: any unexpected enabled?\n- Cron jobs: any unexpected entries?\n\n**Identity & File Security**\n- SOUL.md + AGENTS.md: must be root-owned, 444\n- MEMORY.md, USER.md, cron/jobs.json, openclaw.json: must be 600\n- AGENT_PROMPT.md (this file): must be 600 — your detection playbook must not be world-readable\n- LaunchAgent plists (ai.openclaw.*, com.openclaw.*): must be 600\n- Plaintext key scan: `grep -r \"sk-\\|xai-\\|Bearer\\|api_key\" ~/.openclaw/workspace/ --include=\"*.md\" --include=\"*.json\" --include=\"*.txt\" 2>/dev/null | grep -v \".git\"`\n- .env files in workspace root?\n- Git history secrets: `git -C ~/.openclaw/workspace log -p --all 2>/dev/null | grep -i \"password\\|secret\\|api_key\\|token\" | head -20`\n\n**API Key Handling**\n- Keys in Keychain or flat config files?\n- Keys visible in env vars that could be logged?\n- Plaintext keys in openclaw.json?\n- Check ~/.zshrc for hardcoded secrets\n\n**Agent Behavior**\n- Memory injection scan: `grep -r \"ignore previous\\|new instructions\\|system:\\|you are now\" ~/.openclaw/workspace/memory/ 2>/dev/null`\n- Sub-agents scoped correctly?\n- Any agent with unexpected permissions?\n\n**Dependencies**\n- `brew outdated` — flag openclaw, ollama, node specifically\n- `npm outdated -g` — flag openclaw\n\n### Scoring\n- CRITICAL: immediate fix required, active risk\n- HIGH: fix before going live\n- MEDIUM: fix within 30 days\n- LOW: best practice, fix when convenient\n- PASS: clean\n\n**Security Grade:**\n- A: 0 CRITICAL, 0 HIGH\n- B: 0 CRITICAL, 1–2 HIGH\n- C: 1 CRITICAL or 3+ HIGH\n- D: 2+ CRITICAL\n- F: Active compromise indicators\n\n### Output\n\n**1. Technical Report** (internal)\nFull format — exact commands run, evidence captured, precise remediation steps.\nSave to: `memory/audits/abaddon-YYYY-MM-DD.md`\n\n**2. Summary** (Telegram Security topic)\nLetter grade + top findings. CRITICAL findings → immediate DM alert.","readmeExcerpt":"Skill: Abaddon Owner: enochosbot-bot Summary: Red team security mode for OpenClaw. Runs an adversarial audit on demand or nightly — checks exposed ports, credential leaks, file permissions, suspicious pr... Tags: latest:1.0.0 Version history: v1.0.0 | 2026-02-20T04:14:19.066Z | user Initial release. Red team / adversarial security mode for OpenClaw. Nightly + on-demand. macOS. Letter grade output. Archive index: Arch","codeSnippets":[],"executableExamples":[{"language":"bash","snippet":"cat skills/abaddon/templates/abaddon-prompt.md >> ~/.openclaw/workspace/agents/observer/AGENT_PROMPT.md"},{"language":"bash","snippet":"cp skills/abaddon/templates/abaddon-prompt.md ~/.openclaw/workspace/agents/abaddon/AGENT_PROMPT.md"},{"language":"bash","snippet":"bash skills/abaddon/setup/cron-seed.sh"},{"language":"bash","snippet":"chmod 600 ~/.openclaw/workspace/agents/observer/AGENT_PROMPT.md"}],"parameters":null,"dependencies":[],"permissions":[],"extractedFiles":[{"path":"SKILL.md","content":"---\nname: abaddon\ndescription: Red team security mode for OpenClaw. Runs an adversarial audit on demand or nightly — checks exposed ports, credential leaks, file permissions, suspicious processes, and OpenClaw config posture. Assigns a letter grade. Built for macOS deployments.\n---\n\n# Abaddon ⚔️\n\nMost security audits are defensive — they check what you've locked down. Abaddon runs the other direction. It thinks like an attacker. It looks for what an adversary would find, not just what you remember to check.\n\nOn demand or every night at 3:45 AM. Letter grade every time.\n\n## What It Checks\n\n**Network & Exposure**\n- Listening ports — anything bound to 0.0.0.0 is flagged\n- Gateway binding — should be loopback only\n- SSH Remote Login state\n- Active tunnels (ngrok, cloudflared, unexpected remote access)\n- Firewall and stealth mode\n\n**System Integrity**\n- SIP, FileVault, Gatekeeper\n- macOS version + pending updates\n- XProtect / MRT definitions age\n\n**OpenClaw Configuration**\n- Exec security mode (full / allowlist / deny)\n- Gateway auth enabled?\n- Unexpected cron entries\n- Unexpected plugins\n\n**File Permissions**\n- SOUL.md + AGENTS.md: root-owned, 444\n- MEMORY.md, USER.md, AGENT_PROMPT.md, openclaw.json, cron/jobs.json, LaunchAgent plists: 600\n- Flags anything 644 or wider on sensitive paths\n- Plaintext key scan across workspace\n\n**API Key Handling**\n- Keys in Keychain or flat files?\n- Keys leaking through env vars?\n- Secrets in git history?\n- Hardcoded tokens in .zshrc?\n\n**Agent Behavior**\n- Memory injection scan (prompt injection attempts in memory files)\n- Sub-agent scope check\n- Unexpected agent permissions\n\n**Dependencies**\n- Homebrew outdated (flags openclaw, ollama, node)\n- npm global outdated\n\n## Scoring\n\n| Grade | Criteria |\n|-------|----------|\n| A | 0 CRITICAL, 0 HIGH |\n| B | 0 CRITICAL, 1–2 HIGH |\n| C | 1 CRITICAL or 3+ HIGH |\n| D | 2+ CRITICAL |\n| F | Active compromise indicators |\n\n## Installation\n\n### Step 1 — Copy the Abaddon prompt into your agent\n\nIf you have Gideon (the OpenClaw observer agent), append the red team section:\n\n```bash\ncat skills/abaddon/templates/abaddon-prompt.md >> ~/.openclaw/workspace/agents/observer/AGENT_PROMPT.md\n```\n\nIf you don't have Gideon, use the standalone agent prompt:\n\n```bash\ncp skills/abaddon/templates/abaddon-prompt.md ~/.openclaw/workspace/agents/abaddon/AGENT_PROMPT.md\n```\n\n### Step 2 — Add the nightly cron\n\n```bash\nbash skills/abaddon/setup/cron-seed.sh\n```\n\nThis adds a 3:45 AM CST cron job to `~/.openclaw/cron/jobs.json`. Delivers to Telegram Security topic if configured.\n\n### Step 3 — Lock the agent prompt\n\n```bash\nchmod 600 ~/.openclaw/workspace/agents/observer/AGENT_PROMPT.md\n```\n\nYour detection playbook should never be world-readable.\n\n## Usage\n\n**Manual trigger** — say any of:\n- \"run red team\"\n- \"run Abaddon\"\n- \"run full assessment\"\n- \"Abaddon report\"\n\n**Nightly** — fires automatically at 3:45 AM CST after the standard defensive audit (3:30 AM).\n\n## Output\n\nEvery run produces two things:\n\n1. **"},{"path":"_meta.json","content":"{\n  \"ownerId\": \"kn78z941f26sn68a4s3kcetssh81gk44\",\n  \"slug\": \"abaddon\",\n  \"version\": \"1.0.0\",\n  \"publishedAt\": 1771560859066\n}"},{"path":"SPEC.md","content":"# Abaddon — Red Team Mode (merged into Gideon)\n**Date:** 2026-02-19 | **Status:** Merged into agents/observer/AGENT_PROMPT.md\n**Updated:** 2026-02-19 (removed random-delay trigger, added fixed nightly cron)\n\n---\n\n## What It Is\n\nAbaddon is not a separate agent. It is Gideon's adversarial mode — an offensive red team scan that runs alongside the standard nightly defensive audit.\n\nNamed after the angel of the abyss. Assumes breach. Checks what an attacker would exploit.\n\n---\n\n## Two Modes\n\n| Mode | Schedule | Posture | Output |\n|------|----------|---------|--------|\n| Gideon nightly (defensive) | 3:30 AM CST daily | Passive monitor, internal audit | memory/audits/YYYY-MM-DD.md |\n| Abaddon nightly (offensive) | 3:45 AM CST daily | Adversarial red team | memory/audits/abaddon-YYYY-MM-DD.md |\n| Abaddon manual | On demand | Adversarial red team | Same, plus letter grade posted to Security topic |\n\n---\n\n## Trigger Phrases (manual)\n\n- \"run red team\"\n- \"run Abaddon\"\n- \"run full assessment\"\n- \"Abaddon report\"\n\n---\n\n## What Abaddon Checks\n\n1. Exposed ports and unexpected network listeners\n2. Credential leaks — environment variables, log files, temp files\n3. Process list — suspicious or unexpected processes\n4. File permissions on sensitive paths (openclaw.json, SOUL.md, AGENT_PROMPT.md, LaunchAgent plists)\n5. Unauthorized SSH keys, cron entries, LaunchAgents\n6. ops/exec-audit.log — anomalous command patterns\n7. ops/identity-change-audit.log — rapid/unexpected config modifications\n8. Gateway binding — verify loopback-only\n9. World-readable files in sensitive directories\n10. Assigns a letter grade (A–F)\n\n---\n\n## Scheduling\n\nAbaddon runs as a native OpenClaw cron job in `cron/jobs.json`. No shell script wrapper, no random delay.\n\n**Previous approach (deprecated):** `scripts/abaddon-trigger.sh` with random 0–16hr delay via LaunchAgent. This was removed because the trigger script never actually ran a scan — it only checked gateway health and logged a false \"firing\" message.\n\n**Current approach:** Direct cron entry in jobs.json at 3:45 AM CST, `sessionTarget: isolated`, `wakeMode: now`, `agentTurn` with the red team prompt baked in. Delivers to Telegram Security topic with letter grade.\n\n---\n\n## File Permissions Enforced by Abaddon Findings\n\nThese paths should always be 600 (owner-only). Abaddon flags deviations:\n\n- `~/.openclaw/openclaw.json`\n- `~/.openclaw/cron/jobs.json`\n- `~/.openclaw/workspace/SOUL.md`\n- `~/.openclaw/workspace/MEMORY.md`\n- `~/.openclaw/workspace/USER.md`\n- `~/.openclaw/workspace/agents/observer/AGENT_PROMPT.md`\n- `~/Library/LaunchAgents/ai.openclaw.*.plist`\n- `~/Library/LaunchAgents/com.openclaw.*.plist`\n\nRun `skills/enoch-tuning/setup/lock-identity.sh` to enforce all of the above in one pass.\n\n---\n\n## ClawHub Publishing\n\nWhen published as a standalone skill, Abaddon becomes a distributable red team module. Until then, it lives inside Gideon's AGENT_PROMPT.md.\n\nSee: `agents/observer/AGENT_PROMPT.md` → ## Red Team Mode (Abaddon)"},{"path":"templates/abaddon-prompt.md","content":"---\n\n## Red Team Mode (Abaddon) 🔴\n\nActivated when triggered with: \"run red team\", \"run Abaddon\", \"run full assessment\", or via nightly cron at 3:45 AM.\n\nIn red team mode, shift from passive monitor to active attacker. Think like an adversary. Find what they would find.\n\n### Trigger Detection\nIf the user says \"run red team\", \"run Abaddon\", \"run full assessment\", or \"Abaddon report\" — enter this mode immediately. Do NOT run the standard passive audit. Run the full Abaddon protocol below.\n\n### Attack Vector Checks\n\n**Network & Exposure**\n- `lsof -iTCP -sTCP:LISTEN` — flag anything bound to 0.0.0.0 (should all be 127.0.0.1)\n- Gateway binding: localhost only or exposed to network?\n- SSH Remote Login: `systemsetup -getremotelogin`\n- Active tunnels: any ngrok, cloudflared, or unexpected remote access running?\n- Firewall: enabled? stealth mode on?\n- DNS: any unexpected servers configured?\n\n**System Integrity**\n- SIP: `csrutil status` — must be enabled\n- FileVault: `fdesetup status` — must be On\n- Gatekeeper: `spctl --status` — must be enabled\n- macOS: current version? pending updates?\n\n**OpenClaw Configuration**\n- Exec security mode: full/allowlist/deny — is it appropriate?\n- Auth enabled on gateway?\n- streamMode off? (prevents conversation preview leaks)\n- Plugins: any unexpected enabled?\n- Cron jobs: any unexpected entries?\n\n**Identity & File Security**\n- SOUL.md + AGENTS.md: must be root-owned, 444\n- MEMORY.md, USER.md, cron/jobs.json, openclaw.json: must be 600\n- AGENT_PROMPT.md (this file): must be 600 — your detection playbook must not be world-readable\n- LaunchAgent plists (ai.openclaw.*, com.openclaw.*): must be 600\n- Plaintext key scan: `grep -r \"sk-\\|xai-\\|Bearer\\|api_key\" ~/.openclaw/workspace/ --include=\"*.md\" --include=\"*.json\" --include=\"*.txt\" 2>/dev/null | grep -v \".git\"`\n- .env files in workspace root?\n- Git history secrets: `git -C ~/.openclaw/workspace log -p --all 2>/dev/null | grep -i \"password\\|secret\\|api_key\\|token\" | head -20`\n\n**API Key Handling**\n- Keys in Keychain or flat config files?\n- Keys visible in env vars that could be logged?\n- Plaintext keys in openclaw.json?\n- Check ~/.zshrc for hardcoded secrets\n\n**Agent Behavior**\n- Memory injection scan: `grep -r \"ignore previous\\|new instructions\\|system:\\|you are now\" ~/.openclaw/workspace/memory/ 2>/dev/null`\n- Sub-agents scoped correctly?\n- Any agent with unexpected permissions?\n\n**Dependencies**\n- `brew outdated` — flag openclaw, ollama, node specifically\n- `npm outdated -g` — flag openclaw\n\n### Scoring\n- CRITICAL: immediate fix required, active risk\n- HIGH: fix before going live\n- MEDIUM: fix within 30 days\n- LOW: best practice, fix when convenient\n- PASS: clean\n\n**Security Grade:**\n- A: 0 CRITICAL, 0 HIGH\n- B: 0 CRITICAL, 1–2 HIGH\n- C: 1 CRITICAL or 3+ HIGH\n- D: 2+ CRITICAL\n- F: Active compromise indicators\n\n### Output\n\n**1. Technical Report** (internal)\nFull format — exact commands run, evidence captured, precise remediation steps.\nSave to: `memory/audits/abaddon-YYYY-MM-"}],"languages":[],"docsSourceLabel":"CLAWHUB","editorialOverview":null,"editorialQuality":{"score":100,"threshold":65,"status":"thin","wordCount":1558,"uniquenessScore":44,"reasons":["uniqueness-below-45"]}},"media":{"evidence":{"source":"no-media","verified":false,"confidence":"low","updatedAt":"2026-10-12T02:14:54.338Z","emptyReason":"No screenshots, media assets, or demo links are available."},"primaryImageUrl":null,"mediaAssetCount":0,"assets":[],"demoUrl":null},"ownerResources":{"evidence":{"source":"unclaimed","verified":false,"confidence":"low","updatedAt":"2026-10-12T02:14:54.338Z","emptyReason":"This page has not been claimed by the agent owner."},"hasCustomPage":false,"customPageUpdatedAt":null,"customLinks":[],"structuredLinks":{"docsUrl":null,"demoUrl":null,"supportUrl":null,"pricingUrl":null,"statusUrl":null},"customPage":null},"relatedAgents":{"evidence":{"source":"protocol-neighbors","verified":false,"confidence":"medium","updatedAt":"2026-10-12T14:48:17.291Z","emptyReason":null},"items":[{"id":"8ebccd8e-3863-4187-8355-c3f14e1f9edf","entityType":"agent","canonicalPath":"/agent/iofficeai-aionui","slug":"iofficeai-aionui","name":"AionUi","description":"Free, local, open-source 24/7 Cowork app and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more | 🌟 Star if you like it!","url":"https://github.com/iOfficeAI/AionUi","homepage":"https://www.aionui.com","source":"GITHUB_REPOS","protocols":["MCP","OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-10-09T19:11:12.944Z","createdAt":"2026-02-25T03:38:16.584Z","downloads":null},{"id":"b917f68a-ebff-438e-84f8-3f4b2494c0bc","entityType":"agent","canonicalPath":"/agent/activepieces-activepieces","slug":"activepieces-activepieces","name":"activepieces","description":"AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents","url":"https://github.com/activepieces/activepieces","homepage":"https://www.activepieces.com","source":"GITHUB_REPOS","protocols":["OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-04-15T02:22:12.426Z","createdAt":"2026-02-25T03:38:12.412Z","downloads":null},{"id":"5cb26759-3a39-483f-94cf-276a98c13bb8","entityType":"agent","canonicalPath":"/agent/cherryhq-cherry-studio","slug":"cherryhq-cherry-studio","name":"cherry-studio","description":"AI productivity studio with smart chat, autonomous agents, and 300+ assistants. Unified access to frontier LLMs","url":"https://github.com/CherryHQ/cherry-studio","homepage":"https://cherry-ai.com","source":"GITHUB_REPOS","protocols":["MCP","OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-04-11T14:38:40.986Z","createdAt":"2026-02-25T03:38:19.379Z","downloads":null},{"id":"6f6582d0-5d76-4f0f-b81d-86520247950b","entityType":"agent","canonicalPath":"/agent/copilotkit-copilotkit","slug":"copilotkit-copilotkit","name":"CopilotKit","description":"The Frontend for Agents & Generative UI. React + Angular","url":"https://github.com/CopilotKit/CopilotKit","homepage":"https://docs.copilotkit.ai","source":"GITHUB_REPOS","protocols":["OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-03-25T09:50:57.846Z","createdAt":"2026-02-25T03:39:14.617Z","downloads":null}],"links":{"hub":"/agent","source":"/agent/source/clawhub","protocols":[{"label":"OpenClaw","href":"/agent/protocol/openclew"}]}}}