skills-coach@2.3.1
CAUTION· install onlyPublisher claims: "Explore capability boundaries of a target Skill, analyze optimization potential, generate an optimized version using Training-Free GRPO, and compile results..." 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 · 1 s · 2 scenarios · 0 deviationsStatic findings
'cargo': 'curl --proto \'=https\' --tlsv1.2 -sSf https://sh.rustup.rs | sh',Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.rm -rf /tmp/exec_original_task_XXXFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.result = subprocess.run(Fix: Put hosts, commands and strings in plain source. Decoding at runtime looks identical to malware and costs points on its own.result = subprocess.run(cmd, capture_output=True, text=True)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.'uv': 'Install with: brew install uv OR curl -LsSf https://astral.sh/uv/install.sh | sh',Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.result = subprocess.run(cmd, cwd=sample_agent_dir)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.'cargo': 'Install with: curl --proto \'=https\' --tlsv1.2 -sSf https://sh.rustup.rs | sh',Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.- All subskills operate autonomously without user input between stepsFix: Declare and request only the permissions the task needs.result = subprocess.run(cmd, cwd=detector_dir)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.result = subprocess.run(cmd, cwd=evaluate_agent_dir)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.result = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.result = subprocess.run(cmd, capture_output=True, text=True) # Don't change cwdFix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.result = subprocess.run(cmd, cwd=optimize_agent_dir)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.result = subprocess.run(cmd, cwd=exec_agent_dir)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.process = subprocess.Popen(cmd, cwd=optimize_agent_dir)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.result = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.result = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.'uv': 'Install with: brew install uv OR curl -LsSf https://astral.sh/uv/install.sh | sh',Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.'cargo': 'curl --proto \'=https\' --tlsv1.2 -sSf https://sh.rustup.rs | sh',Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.result = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out."""Monitor the process and kill if stuck."""Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.result = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.result = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.return False, f"Failed to kill stuck process: {e}"Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell."""Monitor a subprocess and kill it if stuck."""Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.result = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.'cargo': 'Install with: curl --proto \'=https\' --tlsv1.2 -sSf https://sh.rustup.rs | sh',Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.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 t1ans1r: 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 totalBuild commands from fixed arguments; do not interpolate untrusted strings into a shell.
- Agent configured for autonomous execution without human approval≈ +4 on the totalDeclare and request only the permissions the task needs.
- Python subprocess execution detected≈ +2 on the totalPut hosts, commands and strings in plain source. Decoding at runtime looks identical to malware and costs points on its own.
- Remote installer piped to shell≈ +2 on the totalBuild commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.