MCP Codex Worker (sorryorc/mcp-codex-worker) is an MCP server listed on the M8ven Trust Index. M8ven has not graded it: we have no way to read this server ourselves. No publisher has claimed this listing.

Back to MCP Trust Indexm8ven
C
Emerging
74/100
29 days ago

MCP Codex Worker

Enables Codex to delegate bulk code reading, patching, and testing to an async worker using cheaper AI models, while receiving compact results.

→ sorryorc/mcp-codex-worker· listed on glama

Emerging. No concerning findings. Grades remain capped until the project builds reputation through adoption. Grades reflect the full trust pyramid: code, verification depth, and reputation. New projects cap at C until adoption is earned.

How we verified

Code Verified⚡ Live Monitored: not connected

Verified is a snapshot. Live keeps it current, and builds your track record.

⚡ Connect GitHub → continuous verification on every pushwhy connect →

Who stands behind it

sorryorc

Source: Glama

Is this your MCP?

Claim it to get a verified publisher badge, a free copy of our full audit findings, and direct contact for any high-priority issues we find. Or connect your repo for our deepest verification, Live Monitored: read-only, revoke anytime. What we access →

⚡ Connect GitHub for Live Monitored

Install from

Glama

M8ven verifies MCPs across every public registry — install directly from whichever one you prefer.

// key findings
✅
No credential exfiltration, no sensitive file access, no obfuscation
Static analysis found nothing flowing your secrets to unexpected places.
✅
Open source with a license and README
Anyone can audit the code, the license is declared, and the publisher documents what it does.
🔐
You'll be asked for 4 credentials: ANTHROPIC_API_KEY, ANTHROPIC_AUTH_TOKEN, FALLBACK_API_KEY, ONEAPI_API_KEY
These are read from process.env at runtime. Make sure you trust where they’ll be sent.
// environment variables
To run this server yourself, you supply these values. They go in your own MCP client configuration and stay on your machine. The secret label means the value is sensitive, not that the server mishandles it.
configADAPTER_ENABLE_THINKING
configADAPTER_FORWARD_MODEL
configADAPTER_HEARTBEAT_MS
configADAPTER_INCLUDE_REASONING
configADAPTER_MAX_RETRIES
configADAPTER_MAX_TOKENS
configADAPTER_STRIP_THINK
configADAPTER_THINKING_MODE
configADAPTER_UPSTREAM_TIMEOUT_MS
configALLOW_BYPASS_PERMISSIONSbypassPermissions is blocked unless explicitly enabled with =1.
configALLOW_OFFICIAL_ANTHROPIC
🔐 secretANTHROPIC_API_KEYONEAPI_API_KEY / Primary gateway key. Keep it out of git.
🔐 secretANTHROPIC_AUTH_TOKEN
configANTHROPIC_BASE_URLONEAPI_BASE_URL / Primary gateway URL.
configANTHROPIC_MODELCLAUDE_MODEL / Real backend model used by the gateway.
configCLAUDE_CLI_MODEL
configCLAUDE_CODE_BARE
configCLAUDE_CODE_COMMAND
configCLAUDE_CODE_DISABLE_NONESSENTIAL_TRAFFIC
configCLAUDE_CODE_MODELModel name passed to Claude Code for local validation, usually sonnet.
configCLAUDE_EFFORT
configCLAUDE_MODEL/ ANTHROPIC_MODEL Real backend model used by the gateway.
configCLAUDE_PERMISSION_MODE
configDISABLE_AUTOUPDATER
configDISABLE_BUG_COMMAND
configDISABLE_ERROR_REPORTING
configDISABLE_TELEMETRY
🔐 secretFALLBACK_API_KEYFALLBACK_BASE_URL / Optional fallback gateway.
configFALLBACK_BASE_URL/ FALLBACK_API_KEY Optional fallback gateway.
configFALLBACK_ESCALATE_MODELFALLBACK_MODELS Optional comma-separated fallback model pool, capped at 3 model candidates for bounded routing. Overrides FALLBACK_MODEL / when set; not a cost cap.
configFALLBACK_MODELFALLBACK_MODELS Optional comma-separated fallback model pool, capped at 3 model candidates for bounded routing. Overrides / FALLBACK_ESCALATE_MODEL when set; not a cost cap.
🔐 secretONEAPI_API_KEYsetx "your-provider-key"
configONEAPI_BASE_URLsetx "https://your-gateway.example.com/v1"
configSANDBOX_ROOTsetx "D:/workspaces"
configUSE_OPENAI_ADAPTER1 to use the local Anthropic-to-OpenAI adapter.
configWORKER_BLOCKING_POLICYWORKER_RELIABILITY_TIER, , WORKER_SEMANTIC_GATE, and WORKER_TOOL_BUDGET record reliability expectations and blocking risk. Defaults are observe-only to avoid surprise stalls.
configWORKER_CATEGORY_ERROR_MAX_PCTGate threshold for category-level tool error rate; default 5. WORKER_TOOL_ERROR_MAX_PCT remains as a legacy alias.
configWORKER_ESCALATE_DIFFICULTY
configWORKER_ESCALATE_MODELupgrades only failed, difficult revise passes.
configWORKER_FALLBACK_WARN_EVERYWarn every N fallback calls; default 5, 0 disables.
configWORKER_ISOLATIONworktree allows safe parallel work in one repo.
configWORKER_LITE_CACHE_DIROptional disk cache directory for lite read-only tools; must be inside SANDBOX_ROOT.
configWORKER_LITE_MODELOptional cheaper model for analyze, review, and failure_digest.
configWORKER_METRICS_FILECost telemetry: writes gateway token usage and worker tool-call categories to JSONL when is set.
configWORKER_METRICS_SHARD_BY_PID
configWORKER_OVERALL_TOOL_ERROR_MAX_PCTGate threshold for total worker tool error rate; default 5, and the rate must stay below this value.
configWORKER_PRICE_TABLEOptional JSON model price overrides for npm run stats; must not affect primary/fallback/escalation selection.
configWORKER_RATIO_GATE
configWORKER_RECEIPT_ARTIFACT_MIN_BYTES
configWORKER_RED_TEAM_MIN_RATIO
configWORKER_RELIABILITY_TIER, WORKER_BLOCKING_POLICY, WORKER_SEMANTIC_GATE, and WORKER_TOOL_BUDGET record reliability expectations and blocking risk. Defaults are observe-only to avoid surprise stalls.
configWORKER_SEARCH_RG_MAX_BUFFER
configWORKER_SEMANTIC_GATEWORKER_RELIABILITY_TIER, WORKER_BLOCKING_POLICY, , and WORKER_TOOL_BUDGET record reliability expectations and blocking risk. Defaults are observe-only to avoid surprise stalls.
configWORKER_SINGLE_TOOL_ERROR_MAX_PCTGate threshold for each individual tool's error rate; default 3, and the rate must stay below this value.
configWORKER_STATS_INFER_LEGACY
configWORKER_STATS_SINCE_MINUTES
configWORKER_TOOL_BUDGETWORKER_RELIABILITY_TIER, WORKER_BLOCKING_POLICY, WORKER_SEMANTIC_GATE, and record reliability expectations and blocking risk. Defaults are observe-only to avoid surprise stalls.
configWORKER_TOOL_CIRCUIT_BREAKERActive runtime containment switch; default enabled. Set 0 to disable per-tool/per-error-class circuits.
configWORKER_TOOL_CIRCUIT_DISABLED
configWORKER_TOOL_CIRCUIT_IMMEDIATE_CLASSESComma-separated classes that open a circuit immediately; default upstream_404,search_timeout,shell_mismatch.
configWORKER_TOOL_CIRCUIT_STATE_FILEOptional persisted circuit state path. Defaults to WORKER_METRICS_FILE + ".state.json" so open circuits and a bounded rolling-event snapshot survive MCP restarts.
configWORKER_TOOL_ERROR_MAX_PCTWORKER_CATEGORY_ERROR_MAX_PCT Gate threshold for category-level tool error rate; default 5. remains as a legacy alias.
configWORKER_TOOL_ERROR_MIN_CALLSMinimum sample count before category or single-tool error-rate gates apply; default 10.
configWORKER_TOOL_REVIEW_DISABLEDSet 1 to disable the runtime review loop.
configWORKER_WORKTREE_DIR
// quality suggestions

Tool annotations

No tools have read-only/destructive annotations

Add readOnlyHint or destructiveHint annotations to every tool so hosts can warn users before invoking.

All four hints declared on every tool

15/15 tools missing one or more hints — start (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); get (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint); get_artifact_slice (missing: readOnlyHint, destructiveHint, idempotentHint, openWorldHint), +12 more. OpenAI's directory rejects tools where any of the four hints are missing or non-boolean.

For every tool, set all four hints (readOnlyHint, destructiveHint, idempotentHint, openWorldHint) to explicit true/false values that match the handler’s actual behaviour.

Secrets not logged

1 secret value sent to console.log

Redact or omit secret values from log output.

Claim the listing to review these findings one by one and send us a correction where you disagree, straight to the team. Claiming also means we tell you when the grade moves, and reach you first if we find anything urgent.

// full audit trail
The findings above are the summary. The full trail, every check we ran, each deduction, the network hosts observed and the dependency advisories, goes to verified publishers, along with an alert whenever a new one lands. Verified publishers can also review each finding and dispute it in one click. Publisher corrections have sharpened several of our checks this month, because the maintainer knows the codebase better than any scanner.
// improvement guidance — verified publishers only
We have 3 concrete improvements we can share with the publisher of this MCP. Each comes with specific guidance to raise the trust score.
// embed badge in your README
[![M8ven Score](https://m8ven.ai/badge/mcp/sorryorc-mcp-codex-worker-1qr3hr)](https://m8ven.ai/mcp/sorryorc-mcp-codex-worker-1qr3hr)
Shows your grade and updates automatically. Prefer no grade? Append ?variant=verified to the badge URL.
commit: dd2d9f0f9ef5b45d19e4a106ed5edd894dd54dbb
code hash: 9da0b614ae2b0e5f081c9c9a6372a63a0d1920ff6ebfd4038c0b0ffc75962c7b
verified: 8/2/2026, 9:23:03 AM
view raw JSON →
Check MCPs from inside your assistant
Tool Check · MCP

Vetting this one by hand? Tool Check is an MCP that scores other MCPs. Add it once and ask Claude, ChatGPT, or any MCP client to grade a server, surface CVEs, check the publisher, and suggest safer alternatives — before you install.

https://m8ven.ai/api/mcp/tool-check
check_toolsearch_toolscompare_toolsrecommend_alternativescheck_publisherreport_concern
How to add it →Free · no account needed · works in any MCP client