context-mode
Flagged as possibly malicious Reviewed: expected behaviour
Bawwab's static rules matched a pattern it treats as critical. Read the review notes: a reviewed false alarm is marked as such.
Why it was flagged
The static rules marked this “Flagged as possibly malicious” because of:
exfil.combo: Reads credentials and makes network calls (critical, 3× · first at server.mjs:28895)
- Kind
- Plugin
- Source
- github
- Repository
- mksglu/context-mode / plugins/claude-code
- Commit
a0e429558100c41fba1bb91be852da23c3423a90- Listed in
- mksglu/context-mode
- Popularity
- 34,325 GitHub stars (2026-11-20)
- Link
- https://github.com/mksglu/context-mode/tree/a0e429558100c41fba1bb91be852da23c3423a90/plugins/claude-code
- Scanned
- 2026-11-20T23:09:37+00:00
- Scan time
- 33.7 s
- Scanned as
- plugin
Review note: A local context-saving MCP server and hooks; its only network code is an npm version check, a git ls-remote tag check, and a URL fetch the user's agent requests, none of which sends secrets, and it downloads no code.
Findings (11)
credential-access: Touches credentials or secrets high
cli.mjs:505
OWASP LLM LLM03 Supply ChainOWASP Agentic ASI03 Identity and Privilege AbuseOWASP Agentic ASI04 Agentic Supply Chain VulnerabilitiesMITRE ATLAS AML.T0055 Unsecured CredentialsMITRE ATLAS AML.T0083 Credentials from AI Agent Configuration
var CLAUDE_USER_CONFIG_FILE = ".claude.json";
What this means: Reads files or environment holding keys, tokens or sessions.
The rule's general advice: Confirm the tool needs this; otherwise do not install. (see the review note for this case)
Expected for what this tool does
Same bundled constant (~/.claude.json) used by the old-install check shared with server.mjs, which only looks for a context-mode entry under mcpServers.
Reviewed by Bawwab maintainers, 2026-11-20
credential-access: Touches credentials or secrets high
server.mjs:28895
OWASP LLM LLM03 Supply ChainOWASP Agentic ASI03 Identity and Privilege AbuseOWASP Agentic ASI04 Agentic Supply Chain VulnerabilitiesMITRE ATLAS AML.T0055 Unsecured CredentialsMITRE ATLAS AML.T0083 Credentials from AI Agent Configuration
var CLAUDE_USER_CONFIG_FILE = ".claude.json";
What this means: Reads files or environment holding keys, tokens or sessions.
The rule's general advice: Confirm the tool needs this; otherwise do not install. (see the review note for this case)
Expected for what this tool does
The constant names ~/.claude.json, which the doctor and ctx_upgrade code opens only to check whether it has a context-mode entry under mcpServers (a presence check for old installs); no contents are returned beyond that, and nothing is sent anywhere.
Reviewed by Bawwab maintainers, 2026-11-20
exfil.combo: Reads credentials and makes network calls critical
server.mjs:28895
OWASP LLM LLM03 Supply ChainOWASP Agentic ASI04 Agentic Supply Chain VulnerabilitiesMITRE ATLAS AML.T0037 Data from Local SystemMITRE ATLAS AML.T0055 Unsecured CredentialsMITRE ATLAS AML.T0025 Exfiltration via Cyber Means
var CLAUDE_USER_CONFIG_FILE = ".claude.json";
What this means: Credential access together with outbound requests in one file is the shape of data theft.
The rule's general advice: Do not install unless the flow is understood and benign. (see the review note for this case)
Known false alarm
The rule paired the .claude.json filename constant with network code elsewhere in the bundle. The file is only checked for a context-mode mcpServers entry. The only outbound requests in the bundle are an npm dist-tags GET, a git ls-remote of github.com/mksglu/context-mode, and the user-directed ctx_index fetch; none carries file contents or credentials.
Reviewed by Bawwab maintainers, 2026-11-20
credential-access: Touches credentials or secrets high
opencode/plugin.js:930
OWASP LLM LLM03 Supply ChainOWASP Agentic ASI03 Identity and Privilege AbuseOWASP Agentic ASI04 Agentic Supply Chain VulnerabilitiesMITRE ATLAS AML.T0055 Unsecured CredentialsMITRE ATLAS AML.T0083 Credentials from AI Agent Configuration
var CLAUDE_USER_CONFIG_FILE = ".claude.json";
What this means: Reads files or environment holding keys, tokens or sessions.
The rule's general advice: Confirm the tool needs this; otherwise do not install. (see the review note for this case)
Expected for what this tool does
Same bundled constant (~/.claude.json) used by the old-install check shared with server.mjs, which only looks for a context-mode entry under mcpServers.
Reviewed by Bawwab maintainers, 2026-11-20
exfil.combo: Reads credentials and makes network calls critical
opencode/plugin.js:930
OWASP LLM LLM03 Supply ChainOWASP Agentic ASI04 Agentic Supply Chain VulnerabilitiesMITRE ATLAS AML.T0037 Data from Local SystemMITRE ATLAS AML.T0055 Unsecured CredentialsMITRE ATLAS AML.T0025 Exfiltration via Cyber Means
var CLAUDE_USER_CONFIG_FILE = ".claude.json";
What this means: Credential access together with outbound requests in one file is the shape of data theft.
The rule's general advice: Do not install unless the flow is understood and benign. (see the review note for this case)
Known false alarm
The rule paired the .claude.json filename constant with network code elsewhere in the bundle. The file is only checked for a context-mode mcpServers entry. The only outbound requests in the bundle are an npm dist-tags GET, a git ls-remote of github.com/mksglu/context-mode, and the user-directed ctx_index fetch; none carries file contents or credentials.
Reviewed by Bawwab maintainers, 2026-11-20
credential-access: Touches credentials or secrets high
pi/index.js:12075
OWASP LLM LLM03 Supply ChainOWASP Agentic ASI03 Identity and Privilege AbuseOWASP Agentic ASI04 Agentic Supply Chain VulnerabilitiesMITRE ATLAS AML.T0055 Unsecured CredentialsMITRE ATLAS AML.T0083 Credentials from AI Agent Configuration
var CLAUDE_USER_CONFIG_FILE = ".claude.json";
What this means: Reads files or environment holding keys, tokens or sessions.
The rule's general advice: Confirm the tool needs this; otherwise do not install. (see the review note for this case)
Expected for what this tool does
Same bundled constant (~/.claude.json) used by the old-install check shared with server.mjs, which only looks for a context-mode entry under mcpServers.
Reviewed by Bawwab maintainers, 2026-11-20
exfil.combo: Reads credentials and makes network calls critical
pi/index.js:12075
OWASP LLM LLM03 Supply ChainOWASP Agentic ASI04 Agentic Supply Chain VulnerabilitiesMITRE ATLAS AML.T0037 Data from Local SystemMITRE ATLAS AML.T0055 Unsecured CredentialsMITRE ATLAS AML.T0025 Exfiltration via Cyber Means
var CLAUDE_USER_CONFIG_FILE = ".claude.json";
What this means: Credential access together with outbound requests in one file is the shape of data theft.
The rule's general advice: Do not install unless the flow is understood and benign. (see the review note for this case)
Known false alarm
The rule paired the .claude.json filename constant with network code elsewhere in the bundle. The file is only checked for a context-mode mcpServers entry. The only outbound requests in the bundle are an npm dist-tags GET, a git ls-remote of github.com/mksglu/context-mode, and the user-directed ctx_index fetch; none carries file contents or credentials.
Reviewed by Bawwab maintainers, 2026-11-20
plugin.hook-broad: Plugin hook runs on every tool call med
.codex-plugin/hooks.json:7
MITRE ATLAS AML.T0142 Event Triggered Execution
node "${PLUGIN_ROOT}/hook.mjs" --agent codex
What this means: A hook with no specific matcher runs a command each time the assistant uses any tool, giving it constant access to your session.
The rule's general advice: Check the hook command, and prefer plugins whose hooks match specific tools. (see the review note for this case)
Expected for what this tool does
The plugin registers hook.mjs for PreToolUse, PostToolUse, SessionStart and UserPromptSubmit so it can route large tool output into its local index; hook.mjs has no network calls. Broad hook coverage is how the tool works.
Reviewed by Bawwab maintainers, 2026-11-20
plugin.hook-broad: Plugin hook runs on every tool call med
.codex-plugin/hooks.json:7
MITRE ATLAS AML.T0142 Event Triggered Execution
node "${PLUGIN_ROOT}/hook.mjs" --agent codex
What this means: A hook with no specific matcher runs a command each time the assistant uses any tool, giving it constant access to your session.
The rule's general advice: Check the hook command, and prefer plugins whose hooks match specific tools. (see the review note for this case)
Expected for what this tool does
The plugin registers hook.mjs for PreToolUse, PostToolUse, SessionStart and UserPromptSubmit so it can route large tool output into its local index; hook.mjs has no network calls. Broad hook coverage is how the tool works.
Reviewed by Bawwab maintainers, 2026-11-20
plugin.hook-broad: Plugin hook runs on every tool call med
.zcode-plugin/hooks.json:7
MITRE ATLAS AML.T0142 Event Triggered Execution
node "${CLAUDE_PLUGIN_ROOT}/hook.mjs" --agent zcode
What this means: A hook with no specific matcher runs a command each time the assistant uses any tool, giving it constant access to your session.
The rule's general advice: Check the hook command, and prefer plugins whose hooks match specific tools. (see the review note for this case)
Expected for what this tool does
The plugin registers hook.mjs for PreToolUse, PostToolUse, SessionStart and UserPromptSubmit so it can route large tool output into its local index; hook.mjs has no network calls. Broad hook coverage is how the tool works.
Reviewed by Bawwab maintainers, 2026-11-20
plugin.hook-broad: Plugin hook runs on every tool call med
.zcode-plugin/hooks.json:7
MITRE ATLAS AML.T0142 Event Triggered Execution
node "${CLAUDE_PLUGIN_ROOT}/hook.mjs" --agent zcode
What this means: A hook with no specific matcher runs a command each time the assistant uses any tool, giving it constant access to your session.
The rule's general advice: Check the hook command, and prefer plugins whose hooks match specific tools. (see the review note for this case)
Expected for what this tool does
The plugin registers hook.mjs for PreToolUse, PostToolUse, SessionStart and UserPromptSubmit so it can route large tool output into its local index; hook.mjs has no network calls. Broad hook coverage is how the tool works.
Reviewed by Bawwab maintainers, 2026-11-20
10 low-severity notes (informational, do not affect the result)
install-dir-access: Reads or writes Claude's install folders low
cli.mjs:316
MITRE ATLAS AML.T0081 Modify AI Agent Configuration
"claude-code": [".claude"],
What this means: Uses ~/.claude/agents, skills, commands or plugins (or a project .claude/ folder). Normal for installers, but it changes what Claude loads.
What you can do: Check which files it adds there and that you want them.
install-dir-access: Reads or writes Claude's install folders low
hook.mjs:2
MITRE ATLAS AML.T0081 Modify AI Agent Configuration
var Xe=["claude-code","codex","gemini","qwen","copilot-cli","vscode","kiro","zcode","pi","opencode","omp"];var te="context-mode",Nt=Object.freeze(["ctx_execute","ctx_batch","ctx_index","ctx_search","c
What this means: Uses ~/.claude/agents, skills, commands or plugins (or a project .claude/ folder). Normal for installers, but it changes what Claude loads.
What you can do: Check which files it adds there and that you want them.
install-dir-access: Reads or writes Claude's install folders low
run.mjs:361
MITRE ATLAS AML.T0081 Modify AI Agent Configuration
"claude-code": [".claude"],
What this means: Uses ~/.claude/agents, skills, commands or plugins (or a project .claude/ folder). Normal for installers, but it changes what Claude loads.
What you can do: Check which files it adds there and that you want them.
install-dir-access: Reads or writes Claude's install folders low
search.mjs:361
MITRE ATLAS AML.T0081 Modify AI Agent Configuration
"claude-code": [".claude"],
What this means: Uses ~/.claude/agents, skills, commands or plugins (or a project .claude/ folder). Normal for installers, but it changes what Claude loads.
What you can do: Check which files it adds there and that you want them.
install-dir-access: Reads or writes Claude's install folders low
server.mjs:28275
MITRE ATLAS AML.T0081 Modify AI Agent Configuration
var HOST_CONFIG_DIR_NAME = ".claude";
What this means: Uses ~/.claude/agents, skills, commands or plugins (or a project .claude/ folder). Normal for installers, but it changes what Claude loads.
What you can do: Check which files it adds there and that you want them.
install-dir-access: Reads or writes Claude's install folders low
opencode/plugin.js:248
MITRE ATLAS AML.T0081 Modify AI Agent Configuration
var HOST_CONFIG_DIR_NAME = ".claude";
What this means: Uses ~/.claude/agents, skills, commands or plugins (or a project .claude/ folder). Normal for installers, but it changes what Claude loads.
What you can do: Check which files it adds there and that you want them.
install-dir-access: Reads or writes Claude's install folders low
pi/index.js:11393
MITRE ATLAS AML.T0081 Modify AI Agent Configuration
var HOST_CONFIG_DIR_NAME = ".claude";
What this means: Uses ~/.claude/agents, skills, commands or plugins (or a project .claude/ folder). Normal for installers, but it changes what Claude loads.
What you can do: Check which files it adds there and that you want them.
plugin.bundled-mcp: Plugin bundles an MCP server low
.claude-plugin/plugin.json
OWASP LLM LLM03 Supply ChainOWASP Agentic ASI04 Agentic Supply Chain VulnerabilitiesMITRE ATLAS AML.T0010.005 Supply Chain Compromise: AI Agent Tool
plugin bundles MCP server(s): context-mode
What this means: The plugin installs additional MCP servers that run code on your machine and are not covered by this layer alone.
What you can do: Review each bundled server as if you were installing it separately.
plugin.bundled-mcp: Plugin bundles an MCP server low
.zcode-plugin/plugin.json
OWASP LLM LLM03 Supply ChainOWASP Agentic ASI04 Agentic Supply Chain VulnerabilitiesMITRE ATLAS AML.T0010.005 Supply Chain Compromise: AI Agent Tool
plugin bundles MCP server(s): context-mode
What this means: The plugin installs additional MCP servers that run code on your machine and are not covered by this layer alone.
What you can do: Review each bundled server as if you were installing it separately.
plugin.bundled-mcp: Plugin bundles an MCP server low
copilot/plugin.json
OWASP LLM LLM03 Supply ChainOWASP Agentic ASI04 Agentic Supply Chain VulnerabilitiesMITRE ATLAS AML.T0010.005 Supply Chain Compromise: AI Agent Tool
plugin bundles MCP server(s): context-mode
What this means: The plugin installs additional MCP servers that run code on your machine and are not covered by this layer alone.
What you can do: Review each bundled server as if you were installing it separately.
Dispute this result
Maintainer and think this is wrong? Open a dispute on GitHub with the target id plugin-mksglu-context-mode-plugins-claude-code-f529ad.