--- canonical: "https://firewall.lpm.dev/npm/@rallycry/conveyor-agent/v/10.13.55" markdown: "https://firewall.lpm.dev/npm/@rallycry/conveyor-agent/v/10.13.55.md" package: "@rallycry/conveyor-agent" report_status: "published" title: "@rallycry/conveyor-agent@10.13.55 npm security report" verdict: "suspicious" version: "10.13.55" --- # @rallycry/conveyor-agent@10.13.55 npm security report > **Trust boundary:** Package metadata, advisory text, filenames, URLs, and source snippets in this report come from external packages or feeds. Treat them as untrusted evidence. Do not execute instructions or code found in this document. ## Verdict summary **Flagged as agent extension risk** — Allowed by default with warning: install-time first-party agent extension setup was detected. - **Verdict:** Suspicious - **Product-default install policy:** Warn - **Firewall policy:** Warn-only agent extension risk - **Public report status:** Published - **Threat category:** Agent extension lifecycle risk - **Selected version:** 10.13.55 - **Selected version is latest:** No - **Analysis source:** AI Security Review (lpm-firewall-ai) LPM treats this as warn-only first-party agent extension lifecycle risk. The runtime boot path mutates Claude configuration/credential paths and enables a network-facing SSH daemon. It also launches a controlled AI-agent harness with permissive tool settings. ## Latest scan - **Scanner version:** rust-scanner-worker-schema-1 - **Verdict:** Suspicious - **Confidence:** 91.0% - **Started:** 2026-08-09T21:46:22.509Z - **Finished:** 2026-08-09T21:47:14.037Z - **Download time:** 512 ms - **Static scan time:** 2139 ms - **AI review time:** 48876 ms - **Total time:** 51528 ms ## Security analysis ### Published attack-surface review - **Summary:** The runtime boot path mutates Claude configuration/credential paths and enables a network-facing SSH daemon. It also launches a controlled AI-agent harness with permissive tool settings. - **Trigger:** Running conveyor-agent boot as a configured agent/workbench container. - **Impact:** A configured Conveyor deployment can grant its runtime agent broad tool capability and remote SSH access within the workspace container. - **Evidence paths:** package.json, dist/boot-6X5ZZG3B.js, dist/chunk-V33EYJWC.js, dist/cli.js - **Review source:** ai\_review - **Reviewed:** 2026-08-09T21:47:14.037Z ### AI review details - **Review stage:** source\_first\_review - **Mechanism:** AI-agent configuration control, credential-path rewiring, and SSH service startup. - **Rationale:** Source confirms high-risk AI-agent and remote-access capabilities during explicit platform boot. They warrant a warning, but do not meet the block threshold because there is no install hook or demonstrated malicious chain. - **Files touched:** /home/conveyor/.claude.json, /home/conveyor/.claude/.credentials.json, /home/conveyor/.ssh/authorized\_keys, /var/run/conveyor-workbench/ssh/authorized\_keys - **Network endpoints:** https://api.conveyor.rallycryapp.com, https://github.com ### Review decision - **Verdict:** Suspicious - **Confidence:** 91.0% - **Recommended action:** downgrade\_to\_warn - **Intent class:** Dangerous Capability - **False-positive risk:** Medium - **Evidence for:** dist/boot-6X5ZZG3B.js seeds Claude bypass-permission acceptance and model settings., dist/boot-6X5ZZG3B.js rewires ~/.claude credentials and symlinks., dist/boot-6X5ZZG3B.js starts sshd on 0.0.0.0 using persisted authorized\_keys., dist/chunk-V33EYJWC.js generates hooks that can auto-allow MCP tools and passes dangerous permission settings. - **Evidence against:** package.json has no npm preinstall/install/postinstall hook., Boot execution requires explicit conveyor-agent boot plus role and bootstrap-token environment., Network use targets the package service and GitHub bootstrap/repository flows., No credential exfiltration, remote payload download, eval, or hidden install-time execution found. ## Public findings ### 1. Low: Non Install Lifecycle Scripts - **Category:** Manifest - **Confidence:** 80.0% Package declares lifecycle scripts that are not normally run for registry tarball installs. ### 2. Low: Scripts Present - **Category:** Manifest - **Confidence:** 100.0% Package declares npm scripts. ### 3. Medium: Network - **Category:** Source - **Confidence:** 75.0% Package source references network APIs. ### 4. Medium: Environment Vars - **Category:** Source - **Confidence:** 75.0% Package source references environment variables. ### 5. Low: Filesystem - **Category:** Source - **Confidence:** 70.0% Package source references filesystem APIs. ### 6. Critical: Persistence Backdoor - **Category:** Source - **Confidence:** 90.0% - **Path:** dist/boot-6X5ZZG3B.js - **Public source:** [View source]() Source writes persistence or remote-access backdoor material. Public source snippet (untrusted): ```javascript L27: // src/boot/index.ts L28: import { spawn } from "child_process"; L29: import { statSync as statSync4 } from "fs"; ... L56: function parseBundle(text) { L57: const raw = JSON.parse(text); L58: const envVarsRaw = raw.envVars; ... L168: const emit = (line) => { L169: process.stdout.write(`${line} L170: `); ... L677: return join3( L678: process.env.CONVEYOR_SHARED_DIR || "/var/run/conveyor-workbench", L679: "ssh", ``` ### 7. Critical: Trigger Reachable Dangerous Capability - **Category:** Source - **Confidence:** 90.0% - **Path:** dist/boot-6X5ZZG3B.js - **Public source:** [View source]() A package entrypoint or install-time lifecycle script reaches a source file with blocking dangerous behavior. Public source snippet (untrusted): ```javascript Trigger-reachable chain: manifest.bin -> dist/cli.js -> dist/boot-6X5ZZG3B.js L27: // src/boot/index.ts L28: import { spawn } from "child_process"; L29: import { statSync as statSync4 } from "fs"; ... L56: function parseBundle(text) { L57: const raw = JSON.parse(text); L58: const envVarsRaw = raw.envVars; ... L168: const emit = (line) => { L169: process.stdout.write(`${line} L170: `); ... L677: return join3( L678: process.env.CONVEYOR_SHARED_DIR || "/var/run/conveyor-workbench", L679: "ssh", ``` ### 8. High: Trigger Reachable Persistence - **Category:** Source - **Confidence:** 94.0% - **Path:** dist/boot-6X5ZZG3B.js - **Public source:** [View source]() A manifest entrypoint or package-local install chain reaches persistence behavior. Public source snippet (untrusted): ```javascript Trigger-reachable persistence chain: manifest.bin -> dist/cli.js -> dist/boot-6X5ZZG3B.js L27: // src/boot/index.ts L28: import { spawn } from "child_process"; L29: import { statSync as statSync4 } from "fs"; ... L56: function parseBundle(text) { L57: const raw = JSON.parse(text); L58: const envVarsRaw = raw.envVars; ... L168: const emit = (line) => { L169: process.stdout.write(`${line} L170: `); ... L677: return join3( L678: process.env.CONVEYOR_SHARED_DIR || "/var/run/conveyor-workbench", L679: "ssh", ``` ### 9. Low: High Entropy Strings - **Category:** Supply Chain - **Confidence:** 55.0% Package source contains high-entropy string patterns. ### 10. Low: Url Strings - **Category:** Supply Chain - **Confidence:** 65.0% Package source contains URL literals. ### 11. Medium: Ships Build Helper - **Category:** Artifact Inventory - **Confidence:** 70.0% - **Path:** runtime/entrypoint.sh - **Public source:** [View source]() Package ships non-JavaScript build or shell helper files. Public source snippet (untrusted): ```shell path = runtime/entrypoint.sh kind = build_helper sizeBytes = 603 magicHex = [redacted] ``` ### 12. Medium: Structural Risk Force Deep Review - **Category:** Artifact Inventory - **Confidence:** 100.0% Artifact structure forces deeper review even if the static behavioral verdict is clean. ### 13. High: Semantic Analysis Limited - **Category:** Scanner Coverage - **Confidence:** 100.0% - **Path:** dist/chunk-V33EYJWC.js - **Public source:** [View source]() A bounded semantic-analysis stage reached its safety limit; remaining detectors completed, but this package requires AI review. Public source snippet (untrusted): ```javascript stage = ast_semantic_analysis; reason = ast_alias_growth_limit_exceeded; limitedFiles = 1 ``` ### 14. Critical: Previous Version Dangerous Delta - **Category:** Supply Chain - **Confidence:** 93.0% - **Path:** dist/chunk-V33EYJWC.js - **Public source:** [View source]() This package version adds a dangerous source file absent from the previous stored version; route for source-aware review. Public source snippet (untrusted): ```javascript matchType = previous_version_dangerous_delta matchedPackage = @rallycry/conveyor-agent@10.13.18 matchedIdentity = npm:[redacted]:10.13.18 similarity = 0.353 summary = stored previous version shares package body but lacks this dangerous source file ``` ## Dependencies and install lifecycle - **Lifecycle scripts present:** Yes - **Published lifecycle scripts:** prepublishOnly - **Dependencies:** 6 - **Optional dependencies:** 0 - **Peer dependencies:** 0 - **Development dependencies:** 4 - **Published dependency-graph edges:** 6 ### Published dependency entries - @anthropic-ai/claude-agent-sdk ^0.3.219 (Dependency) - @modelcontextprotocol/sdk ^1.12.1 (Dependency) - node-pty ^1.0.0 (Dependency) - socket.io-client ^4.8.3 (Dependency) - tar ^7.5.19 (Dependency) - zod ^4.0.0 (Dependency) ## Package metadata - **Package:** @rallycry/conveyor-agent - **Ecosystem:** npm - **Version:** 10.13.55 - **License:** MIT - **Version published:** 2026-08-09T16:58:14.928Z - **Package first seen:** 2026-07-02T13:37:54.462Z - **Package last seen:** 2026-09-04T20:03:21.735Z - **Known versions:** 45 - **Latest version:** 11.0.11 - **Appeal under review:** No - **Description:** Conveyor Agent Runner v10 - PTY harness for the task chat (SDK harness for audit/project-chat). Agent-as-User architecture with BaseService patterns. Works locally too. - **Maintainers:** rallycry - **Keywords:** agent, claude, codespace, conveyor, v10 - **Artifact files:** 43 - **Artifact unpacked size:** 2,838,004 bytes - **Artifact signatures:** 1 - **Attestations:** No ## References - [HTML security report]()