git-mcp-server@0.2.5
CAUTIONPublisher claims: "Git MCP Server - Unified command-line tool for managing Git repositories across GitHub and GitLab" Observed in sandbox: no network, touches nothing outside the project, writes no memory.
Is [email protected] safe to install?
KillerSkills ran it in the KillerSkills honeypot sandbox (2 scenarios) on 2026-11-01 and scored it 71/100 (Caution). It is not known to be malicious, but it has behaviour or code a security policy should decide on before install. Notable findings: Context shift attempt; Context shift attempt; Instructs the agent to modify its persistent memory or instruction files. Summary: no network, touches nothing outside the project, writes no memory.
Behavior trace
Honeypot run · 1 s · 2 scenarios · 0 deviationst+CategoryObserved action
0.0sprocessexited 1:
0.4sprocessdependency install exit 1
0.4sprocessexec /usr/bin/timeout
0.4sprocessexec .venv/bin/git-mcp
0.4sprocessexited 127: timeout: failed to run command '.venv/bin/git-mcp': No such file or directory
0.4sclaimedno tools/list reply (exit 127)
0.4sprocessscript backend runs setup only
Static findings
CRITICAL
Context shift attemptCLAUDE.md:236 · Context shift attempt — trying to change operational modeFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Context shift attemptCLAUDE.md:248 · Context shift attempt — trying to change operational modeFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01HIGH
Instructs the agent to modify its persistent memory or instruction filesgit_mcp/mcp_server.py:1247 · Instructs the agent to modify its persistent memory or instruction filesFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01MEDIUM
Instructs the agent to modify its persistent memory or instruction filestests/test_codex_integration.py:315 · Instructs the agent to modify its persistent memory or instruction filesFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01HIGH
Instructs the agent to modify its persistent memory or instruction filesgit_mcp/mcp_server.py:1051 · Instructs the agent to modify its persistent memory or instruction filesFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01MEDIUM
Access to sensitive file pathstests/test_codex_integration.py:65 · Access to sensitive file paths — SSH keys, AWS credentials, env files, or system filesFix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST05MEDIUM
Python subprocess execution detectedgit_mcp/mcp_server.py:846 · Python subprocess execution detectedFix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05MEDIUM
Remote installer piped to shellinstall.sh:11 · Remote installer piped to shellFix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST07MEDIUM
Instructs the agent to modify its persistent memory or instruction filestests/test_codex_integration.py:329 · Instructs the agent to modify its persistent memory or instruction filesFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01MEDIUM
Python subprocess execution detectedgit_mcp/mcp_server.py:850 · Python subprocess execution detectedFix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05CRITICAL
Pipe-to-shell patterninstall.sh:11 · Pipe-to-shell pattern — downloads and immediately executes remote codeFix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST06CRITICAL
Context shift attemptREADME.md:379 · Context shift attempt — trying to change operational modeFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Context shift attemptREADME.md:408 · Context shift attempt — trying to change operational modeFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Context shift attemptREADME.md:383 · Context shift attempt — trying to change operational modeFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01MEDIUM
Python subprocess execution detectedgit_mcp/commands/mr.py:28 · Python subprocess execution detectedFix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05CRITICAL
Context shift attemptREADME.md:449 · Context shift attempt — trying to change operational modeFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01What would make this Verified
9 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 pypi: a verified publisher gets +50 on publisher trust. It is free and usually instant with GitHub sign-in.
- Does not do what the description says≈ +5 on the totalIn the sandbox the skill never performed the task it claims (no call to the service it is for). Make sure the happy path runs without extra setup, or describe the real behavior.
- New publisher≈ +3 on the totalTime clears this on its own after 14 and 30 days. Claiming the handle (+50) is the fastest way past it.
- Access to sensitive file paths≈ +2 on the totalStop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
- Python subprocess execution detected≈ +2 on the totalBuild commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
Publish a new version to PyPI, then queue it here. Not your skill? Claim the handle to get verdict changes by email.Claim pypi