salesforce-development
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, 4× · first at bin/soql-lsp.bundled.js:35)
- Kind
- Plugin
- Source
- github
- Repository
- forcedotcom/sf-skills / plugins/builder/salesforce-development
- Commit
d3088edf8776a3c23f55aadead587eb3903be2d3- Listed in
- anthropics/claude-plugins-official
- Popularity
- 1,175 GitHub stars (2026-11-20) (official, always included)
- Link
- https://github.com/forcedotcom/sf-skills/tree/d3088edf8776a3c23f55aadead587eb3903be2d3/plugins/builder/salesforce-development
- Scanned
- 2026-11-20T23:26:08+00:00
- Scan time
- 304.4 s
- Scanned as
- plugin
Review note: All flags are bundled Salesforce and language-server libraries, docs and tests; the only credential use is the Salesforce CLI library reading your own org login to talk to your org.
Findings (34)
credential-access: Touches credentials or secrets high
bin/soql-lsp.bundled.js:35
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
`&&f++}_&&v.length>0&&C.push(v.length),this._lineOffsets=C}return this._lineOffsets},p.prototype.positionAt=function(C){C=Math.max(Math.min(C,this._content.length),0);var v=this.getLineOffsets(),_=0,f
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)
Known false alarm
soql-lsp.bundled.js is a minified bundle of the vscode-languageserver library for SOQL. The 'token' hits are LSP progress/cancellation tokens, and the file has no credential file paths or outside hosts. Nothing reads secrets and sends them out.
Reviewed by Claude (AI-assisted review), 2026-11-21
exfil.combo: Reads credentials and makes network calls critical
bin/soql-lsp.bundled.js:35
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
`&&f++}_&&v.length>0&&C.push(v.length),this._lineOffsets=C}return this._lineOffsets},p.prototype.positionAt=function(C){C=Math.max(Math.min(C,this._content.length),0);var v=this.getLineOffsets(),_=0,f
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
soql-lsp.bundled.js is a minified bundle of the vscode-languageserver library for SOQL. The 'token' hits are LSP progress/cancellation tokens, and the file has no credential file paths or outside hosts. Nothing reads secrets and sends them out.
Reviewed by Claude (AI-assisted review), 2026-11-21
files.truncated: Scan limit reached med
scripts/codex_command_surface.py
call-analysis memory budget for this scan exhausted; regex rules only
What this means: Some files, or the tail of a very large file, were not scanned, so malicious code could hide in them.
The rule's general advice: Review the unscanned files manually or raise the limits. (see the review note for this case)
Expected for what this tool does
The scan budget ran out before this file. Checked by hand: a 57-line script that prints the plugin's status banner from local project files. It makes no network calls and reads no secrets.
Reviewed by Claude (AI-assisted review), 2026-11-21
credential-access: Touches credentials or secrets high
scripts/sf-mcp-proxy.bundled.js:40
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
GFS4: `),console.error(e)});Ht[nn]||(YU=global[nn]||[],ez(Ht,YU),Ht.close=function(e){function t(r,n){return e.call(Ht,r,function(i){i||QU(),typeof n=="function"&&n.apply(this,arguments)})}return Obje
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
sf-mcp-proxy.bundled.js and sf-org-info.bundled.js bundle @salesforce/core and jsforce. That library reads the Salesforce CLI's saved org logins (~/.sfdx / ~/.sf, key kept in the OS keychain) to call your own Salesforce org, which is what this plugin is for. The flagged line itself is the bundled graceful-fs module.
Reviewed by Claude (AI-assisted review), 2026-11-21
exfil.host: References an exfiltration-style host high
scripts/sf-mcp-proxy.bundled.js:110
OWASP LLM LLM03 Supply ChainOWASP Agentic ASI04 Agentic Supply Chain VulnerabilitiesMITRE ATLAS AML.T0025 Exfiltration via Cyber Means
<state>${t}</state>
What this means: Paste, webhook and tunnel services are commonly used to receive stolen data.
The rule's general advice: Remove or justify the destination. (see the review note for this case)
Known false alarm
'<state>${t}</state>' is part of the XML body jsforce builds for the Salesforce Bulk API (setting a job's state). It is not a paste, webhook or tunnel host.
Reviewed by Claude (AI-assisted review), 2026-11-21
exfil.combo: Reads credentials and makes network calls med
scripts/sf-mcp-proxy.bundled.js:40
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
GFS4: `),console.error(e)});Ht[nn]||(YU=global[nn]||[],ez(Ht,YU),Ht.close=function(e){function t(r,n){return e.call(Ht,r,function(i){i||QU(),typeof n=="function"&&n.apply(this,arguments)})}return Obje
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)
Expected for what this tool does
The network calls in these bundles go to Salesforce (login.salesforce.com, your org's instance URL, developer.salesforce.com) using the org login the Salesforce CLI already saved. That is the plugin's stated job. No unrelated destination appears in the bundle.
Reviewed by Claude (AI-assisted review), 2026-11-21
credential-access: Touches credentials or secrets high
scripts/sf-org-info.bundled.js:40
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
GFS4: `),console.error(e)});Ht[tn]||(zU=global[tn]||[],BU(Ht,zU),Ht.close=function(e){function t(r,n){return e.call(Ht,r,function(i){i||qU(),typeof n=="function"&&n.apply(this,arguments)})}return Obje
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
sf-mcp-proxy.bundled.js and sf-org-info.bundled.js bundle @salesforce/core and jsforce. That library reads the Salesforce CLI's saved org logins (~/.sfdx / ~/.sf, key kept in the OS keychain) to call your own Salesforce org, which is what this plugin is for. The flagged line itself is the bundled graceful-fs module.
Reviewed by Claude (AI-assisted review), 2026-11-21
exfil.host: References an exfiltration-style host high
scripts/sf-org-info.bundled.js:110
OWASP LLM LLM03 Supply ChainOWASP Agentic ASI04 Agentic Supply Chain VulnerabilitiesMITRE ATLAS AML.T0025 Exfiltration via Cyber Means
<state>${t}</state>
What this means: Paste, webhook and tunnel services are commonly used to receive stolen data.
The rule's general advice: Remove or justify the destination. (see the review note for this case)
Known false alarm
'<state>${t}</state>' is part of the XML body jsforce builds for the Salesforce Bulk API (setting a job's state). It is not a paste, webhook or tunnel host.
Reviewed by Claude (AI-assisted review), 2026-11-21
exfil.combo: Reads credentials and makes network calls med
scripts/sf-org-info.bundled.js:40
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
GFS4: `),console.error(e)});Ht[tn]||(zU=global[tn]||[],BU(Ht,zU),Ht.close=function(e){function t(r,n){return e.call(Ht,r,function(i){i||qU(),typeof n=="function"&&n.apply(this,arguments)})}return Obje
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)
Expected for what this tool does
The network calls in these bundles go to Salesforce (login.salesforce.com, your org's instance URL, developer.salesforce.com) using the org login the Salesforce CLI already saved. That is the plugin's stated job. No unrelated destination appears in the bundle.
Reviewed by Claude (AI-assisted review), 2026-11-21
credential-access: Touches credentials or secrets high
scripts/test/test_discovery_runtime.py:119
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
`~/.claude/settings.json` happens to say on the machine running the test —
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)
Known false alarm
A test docstring explaining that the test patches out plugin detection so it does not depend on what ~/.claude/settings.json says. Nothing reads that file here.
Reviewed by Claude (AI-assisted review), 2026-11-21
credential-access: Touches credentials or secrets high
skills/platform-data-and-tooling-api-context-get/assets/enterprise_api/MobileSecurityPolicy.json:79
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
"description": "The type of mobile security policy. Possible values are: AllowedDeviceList—Allowed Device List Block3dTouch—Block 3D Touch BlockCalendar—Block Calendar BlockCamera—Block Camera BlockCo
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)
Known false alarm
A Salesforce API schema description that lists mobile security policy values (BlockCamera, BlockContacts and so on). It is reference data, not code.
Reviewed by Claude (AI-assisted review), 2026-11-21
files.truncated: Scan limit reached med
vendor/apex-ls/dist/server.node.js:35
environment-flow analysis budget exceeded
What this means: Some files, or the tail of a very large file, were not scanned, so malicious code could hide in them.
The rule's general advice: Review the unscanned files manually or raise the limits. (see the review note for this case)
Expected for what this tool does
These are the vendored Apex language server bundles (about 9 MB each), too large to scan fully. Checked by hand: no credential file paths, and the only URLs are github.com and json-schema.org references.
Reviewed by Claude (AI-assisted review), 2026-11-21
credential-access: Touches credentials or secrets high
vendor/apex-ls/dist/server.node.js:35
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
`&&J++}de&&Oe.length>0&&Me.push(Oe.length),this._lineOffsets=Me}return this._lineOffsets},re.prototype.positionAt=function(Me){Me=Math.max(Math.min(Me,this._content.length),0);var Oe=this.getLineOffse
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)
Known false alarm
Minified bundle of the vendored Apex language server (vscode-languageserver text-document code on the flagged line). It has no credential paths or outside hosts, so the credential and network pairing is between unrelated library parts.
Reviewed by Claude (AI-assisted review), 2026-11-21
exfil.combo: Reads credentials and makes network calls critical
vendor/apex-ls/dist/server.node.js:35
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
`&&J++}de&&Oe.length>0&&Me.push(Oe.length),this._lineOffsets=Me}return this._lineOffsets},re.prototype.positionAt=function(Me){Me=Math.max(Math.min(Me,this._content.length),0);var Oe=this.getLineOffse
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
Minified bundle of the vendored Apex language server (vscode-languageserver text-document code on the flagged line). It has no credential paths or outside hosts, so the credential and network pairing is between unrelated library parts.
Reviewed by Claude (AI-assisted review), 2026-11-21
files.truncated: Scan limit reached med
vendor/apex-ls/dist/worker.platform.js:299
environment-flow analysis budget exceeded
What this means: Some files, or the tail of a very large file, were not scanned, so malicious code could hide in them.
The rule's general advice: Review the unscanned files manually or raise the limits. (see the review note for this case)
Expected for what this tool does
These are the vendored Apex language server bundles (about 9 MB each), too large to scan fully. Checked by hand: no credential file paths, and the only URLs are github.com and json-schema.org references.
Reviewed by Claude (AI-assisted review), 2026-11-21
credential-access: Touches credentials or secrets high
vendor/apex-ls/dist/worker.platform.js:299
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
`,bt==="\r"&&J+1<Le.length&&Le.charAt(J+1)===`
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)
Known false alarm
Minified bundle of the vendored Apex language server (vscode-languageserver text-document code on the flagged line). It has no credential paths or outside hosts, so the credential and network pairing is between unrelated library parts.
Reviewed by Claude (AI-assisted review), 2026-11-21
exfil.combo: Reads credentials and makes network calls critical
vendor/apex-ls/dist/worker.platform.js:299
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
`,bt==="\r"&&J+1<Le.length&&Le.charAt(J+1)===`
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
Minified bundle of the vendored Apex language server (vscode-languageserver text-document code on the flagged line). It has no credential paths or outside hosts, so the credential and network pairing is between unrelated library parts.
Reviewed by Claude (AI-assisted review), 2026-11-21
files.truncated: Scan limit reached med
vendor/apex-ls/dist/worker.platform.js
call-analysis memory budget for this scan exhausted; regex rules only
What this means: Some files, or the tail of a very large file, were not scanned, so malicious code could hide in them.
The rule's general advice: Review the unscanned files manually or raise the limits. (see the review note for this case)
Expected for what this tool does
These are the vendored Apex language server bundles (about 9 MB each), too large to scan fully. Checked by hand: no credential file paths, and the only URLs are github.com and json-schema.org references.
Reviewed by Claude (AI-assisted review), 2026-11-21
credential-access: Touches credentials or secrets high
sf-mcp-proxy.bundled.js:40
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
[server salesforce-api-context] GFS4: `),console.error(e)});Ht[nn]||(YU=global[nn]||[],ez(Ht,YU),Ht.close=function(e){function t(r,n){return e.call(Ht,r,function(i){i||QU(),typeof n=="function"&&n.apply(this,arguments)})}return Obje
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
sf-mcp-proxy.bundled.js and sf-org-info.bundled.js bundle @salesforce/core and jsforce. That library reads the Salesforce CLI's saved org logins (~/.sfdx / ~/.sf, key kept in the OS keychain) to call your own Salesforce org, which is what this plugin is for. The flagged line itself is the bundled graceful-fs module.
Reviewed by Claude (AI-assisted review), 2026-11-21
exfil.host: References an exfiltration-style host high
sf-mcp-proxy.bundled.js:110
OWASP LLM LLM03 Supply ChainOWASP Agentic ASI04 Agentic Supply Chain VulnerabilitiesMITRE ATLAS AML.T0025 Exfiltration via Cyber Means
[server salesforce-api-context] <state>${t}</state>
What this means: Paste, webhook and tunnel services are commonly used to receive stolen data.
The rule's general advice: Remove or justify the destination. (see the review note for this case)
Known false alarm
'<state>${t}</state>' is part of the XML body jsforce builds for the Salesforce Bulk API (setting a job's state). It is not a paste, webhook or tunnel host.
Reviewed by Claude (AI-assisted review), 2026-11-21
exfil.combo: Reads credentials and makes network calls med
sf-mcp-proxy.bundled.js:40
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
[server salesforce-api-context] GFS4: `),console.error(e)});Ht[nn]||(YU=global[nn]||[],ez(Ht,YU),Ht.close=function(e){function t(r,n){return e.call(Ht,r,function(i){i||QU(),typeof n=="function"&&n.apply(this,arguments)})}return Obje
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)
Expected for what this tool does
The network calls in these bundles go to Salesforce (login.salesforce.com, your org's instance URL, developer.salesforce.com) using the org login the Salesforce CLI already saved. That is the plugin's stated job. No unrelated destination appears in the bundle.
Reviewed by Claude (AI-assisted review), 2026-11-21
credential-access: Touches credentials or secrets high
sf-org-info.bundled.js:40
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
[server salesforce-api-context] GFS4: `),console.error(e)});Ht[tn]||(zU=global[tn]||[],BU(Ht,zU),Ht.close=function(e){function t(r,n){return e.call(Ht,r,function(i){i||qU(),typeof n=="function"&&n.apply(this,arguments)})}return Obje
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
sf-mcp-proxy.bundled.js and sf-org-info.bundled.js bundle @salesforce/core and jsforce. That library reads the Salesforce CLI's saved org logins (~/.sfdx / ~/.sf, key kept in the OS keychain) to call your own Salesforce org, which is what this plugin is for. The flagged line itself is the bundled graceful-fs module.
Reviewed by Claude (AI-assisted review), 2026-11-21
exfil.host: References an exfiltration-style host high
sf-org-info.bundled.js:110
OWASP LLM LLM03 Supply ChainOWASP Agentic ASI04 Agentic Supply Chain VulnerabilitiesMITRE ATLAS AML.T0025 Exfiltration via Cyber Means
[server salesforce-api-context] <state>${t}</state>
What this means: Paste, webhook and tunnel services are commonly used to receive stolen data.
The rule's general advice: Remove or justify the destination. (see the review note for this case)
Known false alarm
'<state>${t}</state>' is part of the XML body jsforce builds for the Salesforce Bulk API (setting a job's state). It is not a paste, webhook or tunnel host.
Reviewed by Claude (AI-assisted review), 2026-11-21
exfil.combo: Reads credentials and makes network calls med
sf-org-info.bundled.js:40
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
[server salesforce-api-context] GFS4: `),console.error(e)});Ht[tn]||(zU=global[tn]||[],BU(Ht,zU),Ht.close=function(e){function t(r,n){return e.call(Ht,r,function(i){i||qU(),typeof n=="function"&&n.apply(this,arguments)})}return Obje
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)
Expected for what this tool does
The network calls in these bundles go to Salesforce (login.salesforce.com, your org's instance URL, developer.salesforce.com) using the org login the Salesforce CLI already saved. That is the plugin's stated job. No unrelated destination appears in the bundle.
Reviewed by Claude (AI-assisted review), 2026-11-21
credential-access: Touches credentials or secrets high
test/test_discovery_runtime.py:119
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
[server salesforce-api-context] `~/.claude/settings.json` happens to say on the machine running the test —
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)
Known false alarm
A test docstring explaining that the test patches out plugin detection so it does not depend on what ~/.claude/settings.json says. Nothing reads that file here.
Reviewed by Claude (AI-assisted review), 2026-11-21
credential-access: Touches credentials or secrets high
sf-mcp-proxy.bundled.js:40
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
[server salesforce-metadata-experts] GFS4: `),console.error(e)});Ht[nn]||(YU=global[nn]||[],ez(Ht,YU),Ht.close=function(e){function t(r,n){return e.call(Ht,r,function(i){i||QU(),typeof n=="function"&&n.apply(this,arguments)})}return Obje
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
sf-mcp-proxy.bundled.js and sf-org-info.bundled.js bundle @salesforce/core and jsforce. That library reads the Salesforce CLI's saved org logins (~/.sfdx / ~/.sf, key kept in the OS keychain) to call your own Salesforce org, which is what this plugin is for. The flagged line itself is the bundled graceful-fs module.
Reviewed by Claude (AI-assisted review), 2026-11-21
exfil.host: References an exfiltration-style host high
sf-mcp-proxy.bundled.js:110
OWASP LLM LLM03 Supply ChainOWASP Agentic ASI04 Agentic Supply Chain VulnerabilitiesMITRE ATLAS AML.T0025 Exfiltration via Cyber Means
[server salesforce-metadata-experts] <state>${t}</state>
What this means: Paste, webhook and tunnel services are commonly used to receive stolen data.
The rule's general advice: Remove or justify the destination. (see the review note for this case)
Known false alarm
'<state>${t}</state>' is part of the XML body jsforce builds for the Salesforce Bulk API (setting a job's state). It is not a paste, webhook or tunnel host.
Reviewed by Claude (AI-assisted review), 2026-11-21
exfil.combo: Reads credentials and makes network calls med
sf-mcp-proxy.bundled.js:40
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
[server salesforce-metadata-experts] GFS4: `),console.error(e)});Ht[nn]||(YU=global[nn]||[],ez(Ht,YU),Ht.close=function(e){function t(r,n){return e.call(Ht,r,function(i){i||QU(),typeof n=="function"&&n.apply(this,arguments)})}return Obje
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)
Expected for what this tool does
The network calls in these bundles go to Salesforce (login.salesforce.com, your org's instance URL, developer.salesforce.com) using the org login the Salesforce CLI already saved. That is the plugin's stated job. No unrelated destination appears in the bundle.
Reviewed by Claude (AI-assisted review), 2026-11-21
credential-access: Touches credentials or secrets high
sf-org-info.bundled.js:40
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
[server salesforce-metadata-experts] GFS4: `),console.error(e)});Ht[tn]||(zU=global[tn]||[],BU(Ht,zU),Ht.close=function(e){function t(r,n){return e.call(Ht,r,function(i){i||qU(),typeof n=="function"&&n.apply(this,arguments)})}return Obje
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
sf-mcp-proxy.bundled.js and sf-org-info.bundled.js bundle @salesforce/core and jsforce. That library reads the Salesforce CLI's saved org logins (~/.sfdx / ~/.sf, key kept in the OS keychain) to call your own Salesforce org, which is what this plugin is for. The flagged line itself is the bundled graceful-fs module.
Reviewed by Claude (AI-assisted review), 2026-11-21
exfil.host: References an exfiltration-style host high
sf-org-info.bundled.js:110
OWASP LLM LLM03 Supply ChainOWASP Agentic ASI04 Agentic Supply Chain VulnerabilitiesMITRE ATLAS AML.T0025 Exfiltration via Cyber Means
[server salesforce-metadata-experts] <state>${t}</state>
What this means: Paste, webhook and tunnel services are commonly used to receive stolen data.
The rule's general advice: Remove or justify the destination. (see the review note for this case)
Known false alarm
'<state>${t}</state>' is part of the XML body jsforce builds for the Salesforce Bulk API (setting a job's state). It is not a paste, webhook or tunnel host.
Reviewed by Claude (AI-assisted review), 2026-11-21
exfil.combo: Reads credentials and makes network calls med
sf-org-info.bundled.js:40
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
[server salesforce-metadata-experts] GFS4: `),console.error(e)});Ht[tn]||(zU=global[tn]||[],BU(Ht,zU),Ht.close=function(e){function t(r,n){return e.call(Ht,r,function(i){i||qU(),typeof n=="function"&&n.apply(this,arguments)})}return Obje
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)
Expected for what this tool does
The network calls in these bundles go to Salesforce (login.salesforce.com, your org's instance URL, developer.salesforce.com) using the org login the Salesforce CLI already saved. That is the plugin's stated job. No unrelated destination appears in the bundle.
Reviewed by Claude (AI-assisted review), 2026-11-21
credential-access: Touches credentials or secrets high
test/test_discovery_runtime.py:119
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
[server salesforce-metadata-experts] `~/.claude/settings.json` happens to say on the machine running the test —
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)
Known false alarm
A test docstring explaining that the test patches out plugin detection so it does not depend on what ~/.claude/settings.json says. Nothing reads that file here.
Reviewed by Claude (AI-assisted review), 2026-11-21
credential-access: Touches credentials or secrets high
soql-lsp.bundled.js:35
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
[server salesforce-lsp] `&&f++}_&&v.length>0&&C.push(v.length),this._lineOffsets=C}return this._lineOffsets},p.prototype.positionAt=function(C){C=Math.max(Math.min(C,this._content.length),0);var v=this.getLineOffsets(),_=0,f
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)
Known false alarm
soql-lsp.bundled.js is a minified bundle of the vscode-languageserver library for SOQL. The 'token' hits are LSP progress/cancellation tokens, and the file has no credential file paths or outside hosts. Nothing reads secrets and sends them out.
Reviewed by Claude (AI-assisted review), 2026-11-21
exfil.combo: Reads credentials and makes network calls critical
soql-lsp.bundled.js:35
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
[server salesforce-lsp] `&&f++}_&&v.length>0&&C.push(v.length),this._lineOffsets=C}return this._lineOffsets},p.prototype.positionAt=function(C){C=Math.max(Math.min(C,this._content.length),0);var v=this.getLineOffsets(),_=0,f
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
soql-lsp.bundled.js is a minified bundle of the vscode-languageserver library for SOQL. The 'token' hits are LSP progress/cancellation tokens, and the file has no credential file paths or outside hosts. Nothing reads secrets and sends them out.
Reviewed by Claude (AI-assisted review), 2026-11-21
20 low-severity notes (informational, do not affect the result)
files.analysis-degraded: Analysis degraded by a memory budget low
bin/soql-lsp.bundled.js
call-analysis memory budget exceeded: some calls analysed unresolved, with capped argument text
What this means: A very large file was analysed with every call kept, but with variable values unresolved and long arguments shortened.
What you can do: No action needed on its own; review the file manually if other findings point at it.
credential-access: Touches credentials or secrets low
scripts/codex_hooks.py:173
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
original = dict(os.environ)
What this means: Reads files or environment holding keys, tokens or sessions.
What you can do: Confirm the tool needs this; otherwise do not install.
credential-access: Touches credentials or secrets low
scripts/codex_mcp_launcher.py:147
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
project = resolve_project_cwd(plugin_root=plugin_root, environment=dict(os.environ))
What this means: Reads files or environment holding keys, tokens or sessions.
What you can do: Confirm the tool needs this; otherwise do not install.
files.analysis-degraded: Analysis degraded by a memory budget low
scripts/sf-mcp-proxy.bundled.js
call-analysis memory budget exceeded: some calls analysed unresolved, with capped argument text
What this means: A very large file was analysed with every call kept, but with variable values unresolved and long arguments shortened.
What you can do: No action needed on its own; review the file manually if other findings point at it.
install-dir-access: Reads or writes Claude's install folders low
scripts/sf_context.py:11412
MITRE ATLAS AML.T0081 Modify AI Agent Configuration
return Path(override) if override else Path.home() / ".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.
credential-access: Touches credentials or secrets low
scripts/test/test_sf_context.py:5722
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
env = {k: v for k, v in os.environ.items()
What this means: Reads files or environment holding keys, tokens or sessions.
What you can do: Confirm the tool needs this; otherwise do not install.
files.analysis-degraded: Analysis degraded by a memory budget low
vendor/apex-ls/dist/server.node.js
call-analysis memory budget exceeded: some calls analysed unresolved, with capped argument text
What this means: A very large file was analysed with every call kept, but with variable values unresolved and long arguments shortened.
What you can do: No action needed on its own; review the file manually if other findings point at it.
plugin.bundled-mcp: Plugin bundles an MCP server low
.mcp.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): salesforce-api-context, salesforce-lsp, salesforce-metadata-experts
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
.codex-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): salesforce-api-context, salesforce-lsp, salesforce-metadata-experts
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.
credential-access: Touches credentials or secrets low
codex_hooks.py:173
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
[server salesforce-api-context] original = dict(os.environ)
What this means: Reads files or environment holding keys, tokens or sessions.
What you can do: Confirm the tool needs this; otherwise do not install.
credential-access: Touches credentials or secrets low
codex_mcp_launcher.py:147
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
[server salesforce-api-context] project = resolve_project_cwd(plugin_root=plugin_root, environment=dict(os.environ))
What this means: Reads files or environment holding keys, tokens or sessions.
What you can do: Confirm the tool needs this; otherwise do not install.
files.analysis-degraded: Analysis degraded by a memory budget low
sf-mcp-proxy.bundled.js
[server salesforce-api-context] call-analysis memory budget exceeded: some calls analysed unresolved, with capped argument text
What this means: A very large file was analysed with every call kept, but with variable values unresolved and long arguments shortened.
What you can do: No action needed on its own; review the file manually if other findings point at it.
install-dir-access: Reads or writes Claude's install folders low
sf_context.py:11412
MITRE ATLAS AML.T0081 Modify AI Agent Configuration
[server salesforce-api-context] return Path(override) if override else Path.home() / ".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.
credential-access: Touches credentials or secrets low
test/test_sf_context.py:5722
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
[server salesforce-api-context] env = {k: v for k, v in os.environ.items()
What this means: Reads files or environment holding keys, tokens or sessions.
What you can do: Confirm the tool needs this; otherwise do not install.
credential-access: Touches credentials or secrets low
codex_hooks.py:173
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
[server salesforce-metadata-experts] original = dict(os.environ)
What this means: Reads files or environment holding keys, tokens or sessions.
What you can do: Confirm the tool needs this; otherwise do not install.
credential-access: Touches credentials or secrets low
codex_mcp_launcher.py:147
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
[server salesforce-metadata-experts] project = resolve_project_cwd(plugin_root=plugin_root, environment=dict(os.environ))
What this means: Reads files or environment holding keys, tokens or sessions.
What you can do: Confirm the tool needs this; otherwise do not install.
files.analysis-degraded: Analysis degraded by a memory budget low
sf-mcp-proxy.bundled.js
[server salesforce-metadata-experts] call-analysis memory budget exceeded: some calls analysed unresolved, with capped argument text
What this means: A very large file was analysed with every call kept, but with variable values unresolved and long arguments shortened.
What you can do: No action needed on its own; review the file manually if other findings point at it.
install-dir-access: Reads or writes Claude's install folders low
sf_context.py:11412
MITRE ATLAS AML.T0081 Modify AI Agent Configuration
[server salesforce-metadata-experts] return Path(override) if override else Path.home() / ".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.
credential-access: Touches credentials or secrets low
test/test_sf_context.py:5722
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
[server salesforce-metadata-experts] env = {k: v for k, v in os.environ.items()
What this means: Reads files or environment holding keys, tokens or sessions.
What you can do: Confirm the tool needs this; otherwise do not install.
files.analysis-degraded: Analysis degraded by a memory budget low
soql-lsp.bundled.js
[server salesforce-lsp] call-analysis memory budget exceeded: some calls analysed unresolved, with capped argument text
What this means: A very large file was analysed with every call kept, but with variable values unresolved and long arguments shortened.
What you can do: No action needed on its own; review the file manually if other findings point at it.
Dispute this result
Maintainer and think this is wrong? Open a dispute on GitHub with the target id plugin-forcedotcom-sf-skills-plugins-builder-salesforce-development-42ccd0.