registry / agency-lang / 0.35.0
agency-lang@0.35.0
The Agency language
AI Security Review
scanned 3d ago · by lpm-firewall-aiThe flagged filesystem, subprocess, and persistence features are tied to explicit package operations. Built-in agent logging can include prompts and tool data, but the package documents this behavior, requires a Statelog API key for remote posts, and exposes an observability override.
Static reason
One or more suspicious static signals were detected.; previous stored version diff introduced dangerous source
Trigger
An explicit skills install or scheduling command, audio conversion, or an agent run with Statelog logging configured.
Impact
These actions may write requested skill or schedule files and, when configured, send agent logs to Statelog. No malicious behavior was identified.
Mechanism
Skills and scheduled tasks are created through user-invoked commands. Agent logs use the configured Statelog service and API key; package documentation describes the logged content and configuration.
AI rationale
The flagged filesystem, subprocess, and persistence features are tied to explicit package operations. Built-in agent logging can include prompts and tool data, but the package documents this behavior, requires a Statelog API key for remote posts, and exposes an observability override.
Evidence
~/.claude/skills/<name>/SKILL.md<cwd>/.claude/skills/<name>/SKILL.md~/Library/LaunchAgents/com.agency.schedule.<name>.plistos.tmpdir()/agency-speech-encode-<uuid>.<format>
Network endpoints1
statelog.adit.io
Decision evidence
public snapshotAI called this Clean at 84.0% confidence as Benign with low false-positive risk.
Evidence for AI clean decision
- The manifest lists development, test, build, and CLI scripts; no install hook is shown in the inspected scripts.
- Skill installation is a dedicated CLI command that requires the user to choose a destination, and its installer requires exactly one target flag.
- The built-in agent configures Statelog logging with a Statelog API key; the logger skips remote POSTs when no API key is provided.
- The observability guide discloses that remote logs include prompts, model replies, and tool arguments and results, and describes the host, project, and API key configuration required to send them.
- Runtime config overrides can set observability, allowing the logging behavior to be disabled.
Evidence against
- Built-in agent contexts set observability to true and configure a Statelog host, so an agent run with the relevant API key sends structured logs by default.
Behavioral surface
ChildProcessCryptoDynamicRequireEnvironmentVarsFilesystemNetwork
Source & flagged code
7 flaggeddist/lib/optimize/optimizerModule.jsView file
32// eslint-disable-next-line no-restricted-syntax -- CLI-layer loading of a user artifact; the bundle path is only known at runtime
L33: const mod = await import(pathToFileURL(out).href);
L34: const factory = mod.default;
Medium
Dynamic Require
Package source references dynamic require/import behavior.
dist/lib/optimize/optimizerModule.jsView on unpkg · L32dist/lib/cli/schedule/backends/launchd.jsView file
1import { execFileSync } from "child_process";
L2: import * as fs from "fs";
...
L6: import renderPlist from "../../../templates/cli/schedule/plist.js";
L7: const PLIST_DIR = path.join(os.homedir(), "Library", "LaunchAgents");
L8: function plistPath(name) {
...
L64: agentDir: path.dirname(entry.agentFile),
L65: path: process.env.PATH ?? "/usr/local/bin:/usr/bin:/bin",
L66: home: os.homedir(),
...
L71: fs.mkdirSync(PLIST_DIR, { recursive: true });
L72: // Unload first if already loaded (launchctl load fails on already-loaded plists)
L73: if (fs.existsSync(dest)) {
Medium
Install Persistence
Source writes installer persistence such as shell profile or service configuration.
dist/lib/cli/schedule/backends/launchd.jsFindings
4 High6 Medium5 Low
HighSandbox Evasion Gated Capabilitydist/lib/stdlib/ffmpeg.js
HighTrigger Reachable External Ai Agent Control Surface Mutationdist/lib/cli/skills.js
HighSemantic Analysis Limiteddist/lib/cli/attach.js#virtual:normalized:round1
HighPrevious Version Dangerous Deltadist/lib/host/node/nodeSubprocess.js
MediumDynamic Requiredist/lib/optimize/optimizerModule.js
MediumNetwork
MediumEnvironment Vars
MediumInstall Persistencedist/lib/cli/schedule/backends/launchd.js
MediumShips Build Helperdist/lib/cli/mlxSpeechServer.py
MediumStructural Risk Force Deep Review
LowScripts Present
LowFilesystem
Low
Affected versions and remediation
This report applies to agency-lang@0.35.0.
See all versions of agency-lang for other recorded verdicts.
- Review the evidence and your use of agency-lang@0.35.0 before allowing it. Restrict the permissions described in this report.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.