fs-mcp@1.47.3
CAUTIONPublisher claims: "A secure MCP filesystem server with Stdio and Web UI modes." Observed in sandbox: no network, touches nothing outside the project, writes no memory.
Registry PyPI (MCP server)Publisher luutuankiet-luutuankiet.ftu2 · joined 3 days agoInstalls 324Last detonated 3d ago
Is [email protected] safe to install?
KillerSkills ran it in the KillerSkills honeypot sandbox (2 scenarios) on 2026-10-31 and scored it 63/100 (Caution). It is not known to be malicious, but it has behaviour or code a security policy should decide on before install. Notable findings: Rule modification attempt; Rule modification attempt; Rule modification attempt. Summary: no network, touches nothing outside the project, writes no memory.
Behavior trace
Honeypot run · 1 s · 2 scenarios · 0 deviationst+CategoryObserved action
0.0sprocessno install scripts in bundle
0.2sprocessdependency install exit 1
0.3sprocessexec /usr/bin/timeout
0.3sprocessexec .venv/bin/fs-mcp
0.3sprocessexited 127: timeout: failed to run command '.venv/bin/fs-mcp': No such file or directory
0.3sclaimedno tools/list reply (exit 127)
0.3sprocessscript backend runs setup only
Static findings
CRITICAL
Rule modification attempt.claude/agents/gsd-executor.md:178 · Rule modification attempt — overwriting existing instructionsFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Rule modification attempt.claude/agents/gsd-executor.md:209 · Rule modification attempt — overwriting existing instructionsFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Rule modification attempt.claude/agents/gsd-executor.md:237 · Rule modification attempt — overwriting existing instructionsFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Fake instruction boundary tag.claude/agents/gsd-planner.md:414 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Rule modification attempt.claude/agents/gsd-executor.md:150 · Rule modification attempt — overwriting existing instructionsFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Fake instruction boundary tag.claude/commands/gsd/audit-milestone.md:24 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Fake instruction boundary tag.claude/commands/gsd/add-todo.md:20 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Fake instruction boundary tag.claude/agents/gsd-planner.md:421 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Fake instruction boundary tag.claude/agents/gsd-planner.md:691 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Fake instruction boundary tag.claude/agents/gsd-planner.md:636 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Fake instruction boundary tag.claude/commands/gsd/check-todos.md:19 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Fake instruction boundary tag.claude/commands/gsd/audit-milestone.md:38 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Fake instruction boundary tag.claude/commands/gsd/add-todo.md:18 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Fake instruction boundary tag.claude/commands/gsd/new-milestone.md:46 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Fake instruction boundary tag.claude/commands/gsd/debug.md:27 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Fake instruction boundary tag.claude/commands/gsd/map-codebase.md:26 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Fake instruction boundary tag.claude/commands/gsd/new-milestone.md:570 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Fake instruction boundary tag.claude/commands/gsd/new-milestone.md:559 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Fake instruction boundary tag.claude/commands/gsd/map-codebase.md:36 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Fake instruction boundary tag.claude/commands/gsd/discuss-phase.md:31 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Fake instruction boundary tag.claude/commands/gsd/new-milestone.md:35 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Fake instruction boundary tag.claude/commands/gsd/check-todos.md:22 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Fake instruction boundary tag.claude/commands/gsd/discuss-phase.md:39 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Fake instruction boundary tag.claude/commands/gsd/execute-phase.md:38 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Fake instruction boundary tag.claude/commands/gsd/execute-phase.md:30 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Fake instruction boundary tag.claude/commands/gsd/debug.md:20 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Fake instruction boundary tag.claude/commands/gsd/pause-work.md:80 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Fake instruction boundary tag.claude/commands/gsd/new-project.md:828 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Fake instruction boundary tag.claude/commands/gsd/new-project.md:818 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Fake instruction boundary tag.claude/commands/gsd/plan-phase.md:439 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Fake instruction boundary tag.claude/commands/gsd/pause-work.md:18 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Fake instruction boundary tag.claude/commands/gsd/pause-work.md:82 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Fake instruction boundary tag.claude/commands/gsd/plan-phase.md:31 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Fake instruction boundary tag.claude/commands/gsd/plan-phase.md:200 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Fake instruction boundary tag.claude/commands/gsd/pause-work.md:16 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Fake instruction boundary tag.claude/commands/gsd/plan-phase.md:188 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Fake instruction boundary tag.claude/commands/gsd/plan-phase.md:41 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01HIGH
Prompt injection attempt.claude/agents/gsd-debugger.md:215 · Prompt injection attempt — memory resetFix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
AST01CRITICAL
Fake instruction boundary tag.claude/get-shit-done/references/tdd.md:52 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01HIGH
Direct script execution from GitHub raw URLsrc/fs_mcp/server.py:478 · Direct script execution from GitHub raw URL — unverified code sourceFix: Pin versions and sources; vendor anything fetched at install time.
AST06CRITICAL
Fake instruction boundary tag.claude/commands/gsd/research-phase.md:30 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01HIGH
Access to sensitive file pathssrc/fs_mcp/server.py:151 · 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.
AST05HIGH
Direct script execution from GitHub raw URLsrc/fs_mcp/server.py:929 · Direct script execution from GitHub raw URL — unverified code sourceFix: Pin versions and sources; vendor anything fetched at install time.
AST06CRITICAL
Fake instruction boundary tag.claude/commands/gsd/verify-work.md:28 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Fake instruction boundary tag.claude/commands/gsd/quick.md:34 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Pipe-to-shell patternsrc/fs_mcp/server.py:478 · 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
Pipe-to-shell patternsrc/fs_mcp/server.py:929 · 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.
AST06HIGH
Access to sensitive file paths.claude/agents/gsd-codebase-mapper.md:89 · 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.
AST05CRITICAL
Fake instruction boundary tag.claude/commands/gsd/quick.md:32 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Pipe-to-shell patternsrc/fs_mcp/utils.py:68 · 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
Fake instruction boundary tag.claude/commands/gsd/research-phase.md:117 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Fake instruction boundary tag.claude/commands/gsd/research-phase.md:26 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Pipe-to-shell patternsrc/fs_mcp/utils.py:65 · 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
Fake instruction boundary tag.claude/get-shit-done/references/tdd.md:56 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Fake instruction boundary tag.claude/commands/gsd/research-phase.md:122 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01CRITICAL
Fake instruction boundary tag.claude/commands/gsd/verify-work.md:35 · Fake instruction boundary tagFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.
AST01MEDIUM
Process kill commandsrc/fs_mcp/__main__.py:106 · Process kill command — agent can terminate processesFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05HIGH
Privileged sudo commandsrc/fs_mcp/utils.py:26 · Privileged sudo command — agent may execute with elevated privilegesFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05MEDIUM
Python subprocess execution detectedsrc/fs_mcp/server.py:368 · 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 shellsrc/fs_mcp/utils.py:65 · Remote installer piped to shellFix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST07MEDIUM
File type combined with upload actionsrc/fs_mcp/server.py:2344 · File type combined with upload action — agent may exfiltrate local data filesFix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.
AST07MEDIUM
Python subprocess execution detectedsrc/fs_mcp/__main__.py:79 · Python subprocess execution detectedFix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05MEDIUM
File type combined with upload action.claude/agents/gsd-debugger.md:347 · File type combined with upload action — agent may exfiltrate local data filesFix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.
AST07MEDIUM
Python subprocess execution detectedsrc/fs_mcp/server.py:477 · Python subprocess execution detectedFix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05MEDIUM
Python subprocess execution detectedsrc/fs_mcp/server.py:493 · Python subprocess execution detectedFix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05HIGH
Direct script execution from GitHub raw URLsrc/fs_mcp/utils.py:68 · Direct script execution from GitHub raw URL — unverified code sourceFix: Pin versions and sources; vendor anything fetched at install time.
AST06MEDIUM
Remote installer piped to shellsrc/fs_mcp/server.py:929 · Remote installer piped to shellFix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST07MEDIUM
Python subprocess execution detectedsrc/fs_mcp/edit_tool.py:854 · Python subprocess execution detectedFix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05HIGH
Auto-update or dynamic fetch configurationsrc/fs_mcp/server.py:998 · Auto-update or dynamic fetch configuration — agent may load new code at runtimeFix: Pin versions and sources; vendor anything fetched at install time.
AST06HIGH
Auto-update or dynamic fetch configurationsrc/fs_mcp/server.py:437 · Auto-update or dynamic fetch configuration — agent may load new code at runtimeFix: Pin versions and sources; vendor anything fetched at install time.
AST06MEDIUM
Python subprocess execution detectedsrc/fs_mcp/server.py:470 · Python subprocess execution detectedFix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05HIGH
Auto-update or dynamic fetch configurationsrc/fs_mcp/server.py:929 · Auto-update or dynamic fetch configuration — agent may load new code at runtimeFix: Pin versions and sources; vendor anything fetched at install time.
AST06MEDIUM
Python subprocess execution detectedsrc/fs_mcp/server.py:417 · Python subprocess execution detectedFix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST05HIGH
Direct script execution from GitHub raw URLsrc/fs_mcp/utils.py:65 · Direct script execution from GitHub raw URL — unverified code sourceFix: Pin versions and sources; vendor anything fetched at install time.
AST06HIGH
Privileged sudo commandsrc/fs_mcp/utils.py:24 · Privileged sudo command — agent may execute with elevated privilegesFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05MEDIUM
npx invocation of non-well-known packagesrc/fs_mcp/server.py:631 · npx invocation of non-well-known package — may execute untrusted codeFix: Pin versions and sources; vendor anything fetched at install time.
AST06HIGH
Privileged sudo commandsrc/fs_mcp/utils.py:25 · Privileged sudo command — agent may execute with elevated privilegesFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05MEDIUM
Python subprocess execution detectedsrc/fs_mcp/__main__.py:56 · 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 shellsrc/fs_mcp/server.py:478 · Remote installer piped to shellFix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST07HIGH
Auto-update or dynamic fetch configurationsrc/fs_mcp/server.py:996 · Auto-update or dynamic fetch configuration — agent may load new code at runtimeFix: Pin versions and sources; vendor anything fetched at install time.
AST06MEDIUM
Remote installer piped to shellsrc/fs_mcp/utils.py:68 · Remote installer piped to shellFix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
AST07MEDIUM
File permission modification commandsrc/fs_mcp/utils.py:39 · File permission modification command — agent changes file ownership/permissionsFix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05MEDIUM
File permission modification commandsrc/fs_mcp/utils.py:55 · File permission modification command — agent changes file ownership/permissionsFix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05MEDIUM
Access to sensitive file pathstests/test_server.py:15 · 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
File permission modification commandsrc/fs_mcp/utils.py:40 · File permission modification command — agent changes file ownership/permissionsFix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05MEDIUM
File permission modification commandsrc/fs_mcp/utils.py:54 · File permission modification command — agent changes file ownership/permissionsFix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05MEDIUM
File permission modification commandsrc/fs_mcp/utils.py:47 · File permission modification command — agent changes file ownership/permissionsFix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.
AST05What would make this Verified
17 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 luutuankiet-luutuankiet.ftu2: 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.
- Direct script execution from GitHub raw URL≈ +4 on the totalPin versions and sources; vendor anything fetched at install time.
- 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.
- Privileged sudo command≈ +4 on the totalBuild commands from fixed arguments; do not interpolate untrusted strings into a shell.
Publish a new version to PyPI, then queue it here. Not your skill? Claim the handle to get verdict changes by email.Claim luutuankiet-luutuankiet.ftu2