evoclaw-evolved@1.1.0
CAUTION· install onlyPublisher claims: "EvoClaw (Evolved) maintains and evolves your agent identity by processing experience through classification, memory, reflection, proposal, and governance cyc..." Observed in sandbox: no network, touches nothing outside the project, writes no memory.
Coverage
Bundled scripts ran with no arguments; no agent task scenario for this bundle.- Tested
- 8 of 22 bundled scripts run with no arguments (the rest were not run: per-bundle cap), with decoy credentials in the environment · every file read, file written and host contacted by what ran
- Not tested
- use of the skill by an agent
Is [email protected] safe to install?
KillerSkills ran it in the KillerSkills honeypot sandbox (2 scenarios; bundled scripts ran with no arguments; no agent task scenario for this bundle) on 2026-11-19 and scored it 70/100 (Caution). It is not known to be malicious, but it has behaviour or code a security policy should decide on before install. Notable findings: Pipe-to-shell pattern; Destructive file system command detected; Pipe-to-shell pattern. Summary: no network, touches nothing outside the project, writes no memory.
Behavior trace
Honeypot run · 2 s · 2 scenarios · 0 deviationsStatic findings
"Bash(curl * | sh)",Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.print(f"Bash(sudo rm -rf /): {ctx.query('Bash(sudo rm -rf /)')}") # denyFix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts."Bash(wget * | sh)",Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out."Bash(chmod *)",Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts."Write(~/.ssh/*)",Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose."Bash(curl * | sh)",Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.@import url('https://fonts.googleapis.com/css2?family=JetBrains+Mono:wght@300;400;500;700&family=DM+Sans:ital,wght@0,300;0,400;0,500;0,700;1,400&display=swap');Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose."Bash(sudo *)",Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.ctx = ctx.add_always_deny("Bash(sudo *)", "危险命令")Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.print(f"Bash(sudo rm -rf /): {ctx.query('Bash(sudo rm -rf /)')}") # denyFix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.### Step 8: Update AGENTS.mdFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.result = subprocess.run(cmd, capture_output=True, text=True, timeout=30)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out."Read(/etc/shadow)",Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.# 实际 kill 由 subagents tool 处理Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.@import url('https://fonts.googleapis.com/css2?family=JetBrains+Mono:wght@300;400;500;700&family=DM+Sans:ital,wght@0,300;0,400;0,500;0,700;1,400&display=swap');Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.# 2. stopTask() 统一入口, 按类型分发 killFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.Add this in the Memory area of AGENTS.md:Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.3. Update `MEMORY.md` with distilled learnings from reflectionsFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.Cost to the agent
Tokens estimated at four characters each. Skills load their instruction text when triggered; the figure is what that costs.
Known vulnerabilities
Vulnerability advisories (OSV) cover npm and PyPI packages; this registry has no advisory database.
What would make this Verified
10 points to 80 · weights: behavior 45%, static 25%, prompt 15%, publisher 15%- Pipe-to-shell pattern≈ +9 on the totalBuild commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
- Unclaimed publisher≈ +8 on the totalClaim shuoyu017: a verified publisher gets +50 on publisher trust. It is free and usually instant with GitHub sign-in.
- Destructive file system command detected≈ +4 on the totalRequest the narrowest permission that works; replace wildcards with the exact paths or hosts.
- Access to sensitive file paths≈ +4 on the totalStop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
- URL with embedded credentials≈ +4 on the totalStop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
- Privileged sudo command≈ +4 on the totalRequest the narrowest permission that works; replace wildcards with the exact paths or hosts.