SKILLEMALL.ai
GitHub ↗ RU Sign in

FB totalreclaw

End-to-end encrypted memory for AI agents — portable, yours forever. AES-256-GCM E2EE: server never sees plaintext.

Not recommendedlow grade F
LeoYeAI/openclaw-master-skills Agent Skills author: LeoYeAI MIT 80 files · 1 script body ≈ 7 190 tokens Open the sourcegithub.com↗ analyzed 2 d ago

End-to-end encrypted memory for AI agents — portable, yours forever.

As a process B 65/100 · Nearly there — weak spots: result and completion, inputs and preconditions, progress reporting

IntegrationPostgreSQLAI and agentstype and topics are labelled automatically from the skill text
Runs in: OpenClaw Claude / Claude Code ChatGPT / Codex Gemini CLI Cursor GitHub Copilot / VS Code Qwen Code Kimi Code CLI Z.ai (GLM) DeepSeek (Deep Code) Mistral Vibe LangChain (deepagents) Goose OpenCode
JSON
Technical rating
F
39/100
safety, quality, tests
Safety 60%
15
Quality 40%
75
Run on models
none yet
Process rating
B
65/100
Nearly there
Result and completion w 14
0
Inputs and preconditions w 11
0
Progress reporting w 2
0
the three weakest of ten parameters · all ten

What is at stake

Medium-severity findings: the skill is probably honest, but read what alarmed the scanner.

Risky intent medium severity

Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.

If you install

The purpose itself is risky: wallets, browser password stores, offensive security. Even an honest implementation gives the agent access to things that cost money.

For the author

Explain in the description why the access is needed and how it is limited; add tests that show refusals on dangerous requests.

Dangerous commands medium severity

Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.

If you install

The skill contains commands that delete files, rewrite disks or execute code fetched from the network. The agent may run them without asking if it believes the instructions require it.

For the author

Replace destructive commands with safe equivalents that ask for confirmation, scope them to one folder, and stop piping curl into a shell: pin a version and a checksum.

How to improve

  1. The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
For the model run — optional
  • Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
  • A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.

Guard findings · 29

✓ No critical or high findings

Medium and low: 29
  • medium Risky intent intent-wallet-secrets plugin/crypto.ts:92
    Handles crypto-wallet secrets (seed / mnemonic / private key) — a classic stealer target
    * Derive auth, encryption, and dedup keys from a recovery phrase.
  • medium Risky intent intent-wallet-secrets plugin/index.ts:332
    Handles crypto-wallet secrets (seed / mnemonic / private key) — a classic stealer target
    /** True when recovery phrase is missing — tools return setup instructions. */
  • medium Risky intent intent-wallet-secrets plugin/index.ts:336
    Handles crypto-wallet secrets (seed / mnemonic / private key) — a classic stealer target
    * Derive keys from the recovery phrase, load or create credentials, and
  • medium Risky intent intent-wallet-secrets plugin/index.ts:430
    Handles crypto-wallet secrets (seed / mnemonic / private key) — a classic stealer target
    // One-time billing check for returning users (imported recovery phrase).
  • medium Dangerous commands cmd-persistence plugin/index.ts:494
    Persistence mechanism (cron / launchd / scheduled task / autorun registry) (string literal in code, not executed)
    '~/Library/LaunchAgents/ai.openclaw.gateway.plist under <key>EnvironmentVariables</key>:\n' +
    code literal
  • medium Risky intent intent-wallet-secrets plugin/index.ts:532
    Handles crypto-wallet secrets (seed / mnemonic / private key) — a classic stealer target
    /** Recovery phrase cached for LSH seed derivation (set during initialize()). */
  • medium Risky intent intent-wallet-secrets plugin/README.md:39
    Handles crypto-wallet secrets (seed / mnemonic / private key) — a classic stealer target
    The agent handles setup: generates your encryption keys, asks you to save a 12-word recovery phrase, and registers you. After that, memory is fully automatic.
  • medium Risky intent intent-wallet-secrets plugin/README.md:93
    Handles crypto-wallet secrets (seed / mnemonic / private key) — a classic stealer target
    Same encryption, same recovery phrase, same memories across all agents.
  • medium Risky intent intent-wallet-secrets README.md:247
    Handles crypto-wallet secrets (seed / mnemonic / private key) — a classic stealer target
    | `TOTALRECLAW_RECOVERY_PHRASE` | **Yes** | -- | 12-word BIP-39 recovery phrase (never sent to server) |
  • medium Risky intent intent-wallet-secrets README.md:315
    Handles crypto-wallet secrets (seed / mnemonic / private key) — a classic stealer target
    1. **Key Derivation**: Recovery phrase is processed through Argon2id to derive encryption keys. The phrase is never sent to the server.
  • medium Risky intent intent-wallet-secrets SKILL.md:29
    Handles crypto-wallet secrets (seed / mnemonic / private key) — a classic stealer target
    ## Recovery Phrase -- READ THIS FIRST
  • medium Risky intent intent-wallet-secrets SKILL.md:33
    Handles crypto-wallet secrets (seed / mnemonic / private key) — a classic stealer target
    The recovery phrase is the user's ONLY identity in TotalReclaw. It derives all encryption keys. There is NO password reset, NO recovery mechanism, and NO support channel that can help if it is lost.
  • medium Risky intent intent-wallet-secrets SKILL.md:35
    Handles crypto-wallet secrets (seed / mnemonic / private key) — a classic stealer target
    **When showing the recovery phrase to the user, ALWAYS include this warning:**
  • medium Risky intent intent-wallet-secrets SKILL.md:37
    Handles crypto-wallet secrets (seed / mnemonic / private key) — a classic stealer target
    > Your recovery phrase is the ONLY way to access your encrypted memories. If you lose it, your memories are gone forever -- there is no password reset, no recovery, and no support that can help. Write
  • low Risky intent intent-wallet-secrets CLAWHUB.md:63
    Handles crypto-wallet secrets (seed / mnemonic / private key) — a classic stealer target (quoted — discussed, not commanded)
    description: "End-to-end encrypted memory for AI agents — portable, yours forever. AES-256-GCM E2EE. One recovery phrase, full portability."
    quoted
  • low Risky intent intent-wallet-secrets dist/triggers/before-agent-start.js:296
    Handles crypto-wallet secrets (seed / mnemonic / private key) — a classic stealer target (quoted — discussed, not commanded)
    ? '\n\nTotalReclaw authentication failed. If using a recovery phrase, check that all 12 words are in the correct order and spelled correctly.'
    quoted
  • low Risky intent intent-wallet-secrets plugin/api-client.ts:104
    Handles crypto-wallet secrets (seed / mnemonic / private key) — a classic stealer target (quoted — discussed, not commanded)
    ? ' Authentication failed. If using a recovery phrase, check that all 12 words are in the correct order and spelled correctly.'
    quoted
  • low Risky intent intent-wallet-secrets plugin/generate-mnemonic.ts:11
    Handles crypto-wallet secrets (seed / mnemonic / private key) — a classic stealer target (quoted — discussed, not commanded)
    console.log('\n  Your TotalReclaw recovery phrase (12 words):\n');
    quoted
  • low Risky intent intent-wallet-secrets plugin/index.ts:481
    Handles crypto-wallet secrets (seed / mnemonic / private key) — a classic stealer target (quoted — discussed, not commanded)
    '1. Set TOTALRECLAW_RECOVERY_PHRASE — ask the user if they have an existing recovery phrase or generate a new 12-word BIP-39 mnemonic.\n' +
    quoted
  • low Secrets in code secret-high-entropy-token plugin/package-lock.json:144
    High-entropy token-like string (may be an id, hash or a credential) (detector / deny-list definition)
    "integrity": "sha5…byA+l3XtsAj+Q8tf…oOo+X6HZ…Q8A==",
    detector
  • low Secrets in code secret-high-entropy-token plugin/package-lock.json:192
    High-entropy token-like string (may be an id, hash or a credential) (detector / deny-list definition)
    "integrity": "sha5…bSm+x6ETixtKZBh/qbRE…8Sr/Wcyx…yGA==",
    detector
  • low Secrets in code secret-high-entropy-token plugin/package-lock.json:240
    High-entropy token-like string (may be an id, hash or a credential) (detector / deny-list definition)
    "integrity": "sha5…UzJ+En8KcVm9Lk5+uGUQ…GXw==",
    detector
  • low Secrets in code secret-high-entropy-token plugin/package-lock.json:272
    High-entropy token-like string (may be an id, hash or a credential) (detector / deny-list definition)
    "integrity": "sha5…595+ujv0…pwV/GONa…zQo/1O6zRIkh0m/8+5Bjr…SZw==",
    detector
  • low Secrets in code secret-high-entropy-token plugin/package-lock.json:360
    High-entropy token-like string (may be an id, hash or a credential) (detector / deny-list definition)
    "integrity": "sha5…bUj+rBev…oyX+04WG…4jg==",
    detector
  • low Secrets in code secret-high-entropy-token plugin/subgraph-store.ts:26
    High-entropy token-like string (may be an id, hash or a credential) (quoted — discussed, not commanded)
    const DEFAULT_DATA_EDGE_ADDRESS = '0xC4…5ca';
    quoted
  • low Secrets in code secret-high-entropy-token plugin/subgraph-store.ts:29
    High-entropy token-like string (may be an id, hash or a credential) (quoted — discussed, not commanded)
    const DEFAULT_ENTRYPOINT_ADDRESS = '0x00…032';
    quoted
  • low Risky intent intent-wallet-secrets README.md:27
    Handles crypto-wallet secrets (seed / mnemonic / private key) — a classic stealer target (quoted — discussed, not commanded)
    export TOTALRECLAW_RECOVERY_PHRASE="your twelve word recovery phrase here"
    quoted
  • low Risky intent intent-wallet-secrets README.md:100
    Handles crypto-wallet secrets (seed / mnemonic / private key) — a classic stealer target (quoted — discussed, not commanded)
    export TOTALRECLAW_RECOVERY_PHRASE="your twelve word recovery phrase here"
    quoted
  • low Risky intent intent-wallet-secrets SKILL.md:31
    Handles crypto-wallet secrets (seed / mnemonic / private key) — a classic stealer target (detector / deny-list definition)
    **CRITICAL SAFETY WARNING:** TotalReclaw generates a NEW recovery phrase for the user. NEVER let the user enter a recovery phrase from an existing crypto wallet they own. TotalReclaw uses BIP-39 mnemo
    detector

Files scanned: 80. Evidence is masked. Grey chips explain why severity was lowered.

Against the Agent Skills spec

  • warning body-long SKILL.md body ≈ 7190 tokens (recommended < 5000); move details to references/
  • note frontmatter-key unknown frontmatter key "homepage"

Process rating: all ten parameters 65/100

  • 0Result and completion. Does not say what the result is
  • 0Inputs and preconditions. Does not say what the process needs to start
  • 0Progress reporting. Says nothing while it works
  • 50When it triggers. No condition that starts the skill
  • 70Execution cost. Instruction body is 7190 tokens
  • 100Tools and files. No external tools needed
  • 100Steps. 63 steps
  • 100Failures and branches. 3 branches, has a failure section
  • 100Consistency. Name and required fields are in place
  • 100Running it twice. Mutating operations check current state
  • medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
  • low 11 top-level sections: this looks like several domains in one skill

Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.

Quality signals

  • +5Description has no quoted example phrases that should trigger the skill
  • +3Description length 115: 120–800 characters recommended
  • +3Output format is not stated: the model decides each time
  • +2Single-language instructions
  • +4Description says when NOT to use the skill
  • +4Structure: 48 headings
  • +3Step-by-step instructions: 63 items
  • +4Has examples (34 code blocks)
  • +1License stated

Quality base 70; lint remarks subtract, signals add up to 100. Result: 75.