SKILLEMALL.ai
GitHub ↗ RU Sign in

AC ai-collab

Multi-agent autonomous collaboration system for two OpenClaw agents working in parallel. Use when setting up agent-to-agent communication, running a daemon agent alongside a primary agent, coordinating tasks between Claude and GPT instances, or establishing a shared chat log and inbox protocol. Triggers on: 'set up agent collaboration', 'run two agents', 'agent daemon', 'multi-agent', 'Jim and Clawdy', 'secondary agent', 'agent handoff'.

modbender/skill-library-mcp Agent Skills author: modbender MIT 10 files · 4 scripts body ≈ 3 440 tokens Open the sourcegithub.com↗ analyzed 2 d ago

Multi-agent autonomous collaboration system for two OpenClaw agents working in parallel.

As a process C 57/100 · Has gaps — weak spots: result and completion, inputs and preconditions, running it twice

ProcedureTelegramAI 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
A
90/100
safety, quality, tests
Safety 60%
90
Quality 40%
91
Run on models
none yet
Process rating
C
57/100
Has gaps
Result and completion w 14
0
Inputs and preconditions w 11
0
Running it twice w 4
30
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.

Exfiltration 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 instructions or scripts send data out: environment variables, keys, file contents, chat history. You may never notice, because the agent performs the upload as "part of the task".

For the author

If the upload is not needed for the task, remove it: catalogs flag such skills and delist them. If it is needed, name the destination explicitly, say what leaves the machine, and give the user a switch.

How to improve

    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 · 6

    ✓ No critical or high findings

    Medium and low: 6
    • medium Exfiltration net-credential-use SKILL.md:255
      Credential used in a network call (verify the destination is the intended service)
      UPDATES=$(curl -s "https://api.telegram.org/bot${TELEGRAM_BOT_TOKEN}/getUpdates?offset=…))&timeout=20")
    • low Exfiltration exfil-webhook-url references/telegram-bridge.md:13
      Webhook / callback URL commonly used for exfiltration (verify the destination) (placeholder value)
      curl -s "https://api.telegram.org/bot<BOT_TOKEN>/getUpdates" | python3 -m json.tool | grep '"id"' | head -5
      placeholder
    • low Exfiltration read-dotenv references/telegram-bridge.md:36
      Reads a .env file
      source ~/.openclaw/.env
    • low Exfiltration exfil-webhook-url references/telegram-bridge.md:56
      Webhook / callback URL commonly used for exfiltration (verify the destination) (placeholder value)
      "https://api.telegram.org/bot${BOT_TOKEN}/getUpdates?offset=…))&limit=10&timeout=20&allowed_updates=[\"message\"]")
      placeholder
    • low Exfiltration read-dotenv SKILL.md:251
      Reads a .env file
      source ~/.openclaw/.env
    • low Exfiltration exfil-webhook-url SKILL.md:255
      Webhook / callback URL commonly used for exfiltration (verify the destination) (placeholder value)
      UPDATES=$(curl -s "https://api.telegram.org/bot${TELEGRAM_BOT_TOKEN}/getUpdates?offset=…))&timeout=20")
      placeholder

    Files scanned: 10. Evidence is masked. Grey chips explain why severity was lowered.

    Against the Agent Skills spec

    ✓ No remarks against the Agent Skills spec

    Process rating: all ten parameters 57/100

    • 0Result and completion. Does not say what the result is
    • 0Inputs and preconditions. Does not say what the process needs to start
    • 30Running it twice. 10 mutating operations with no state check
    • 60Tools and files. Uses tools (bash, web) that frontmatter does not declare
    • 60Failures and branches. 2 branches
    • 70When it triggers. States when to use, but not when not to
    • 100Steps. 29 steps
    • 100Consistency. Name and required fields are in place
    • 100Execution cost. Instruction body is 3440 tokens
    • 100Progress reporting. Reports progress
    • medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
    • low 17 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
    • +4Description does not say when NOT to use the skill (false activations)
    • +3Output format is not stated: the model decides each time
    • +1No license
    • +2Single-language instructions
    • +3Description length 441: enough signal without eating the budget
    • +4Structure: 21 headings
    • +3Step-by-step instructions: 29 items
    • +4Has examples (20 code blocks)
    • +4Reference files are cited in the instructions (1 of 2)
    • +3All 4 scripts are documented

    Quality base 70; lint remarks subtract, signals add up to 100. Result: 91.