Files
shockbot/test/utils.ts
T
Colin McDonnell 3a7145db1a Scope installation token permissions in restricted mode (#226)
* Scope installation token permissions in restricted mode

In restricted/disabled bash mode, the installation token is now scoped
to match the workflow's permissions block. This preserves fork push
capability while limiting what the agent can do with the token.

- Read workflow permissions from pullfrog.yml at runtime
- Pass permissions to API when acquiring installation token
- Clear OIDC env vars in restricted mode to prevent token minting
- Simplify setupGit by moving token resolution to main.ts

* Address review feedback: fail closed with default permissions

- Add restrictive default permissions (contents:read, pull_requests:read,
  issues:read) as fallback when workflow permissions can't be read
- Add support for job-level permissions via GITHUB_JOB env var
- Fix misleading comment about token resolution in restricted mode
- Add documentation about fork PR checkout behavior

* Simplify to separate git/MCP tokens without workflow permission scoping

- gitToken: minimal contents:write only (assumed exfiltratable)
- mcpToken: full installation token (not exfiltratable via MCP tools)
- Remove workflowPermissions.ts - security-conscious users can pass
  their own token via GH_TOKEN or inputs.token
- Add type-safe InstallationTokenPermissions to github.ts and API route

* Rename `write` permission to `push` and remove vestigial tool blocking

The `write` permission was previously used to block local file write tools
in agents. This was security theater since bash can write files anyway.

Now `push` only controls the git token scope:
- push: enabled → contents:write (can push commits)
- push: disabled → contents:read (read-only, can't push)

Changes:
- Rename `write` to `push` in action.yml, Prisma schema, and all TS types
- Remove vestigial write tool blocking from all agents (claude, cursor,
  gemini, opencode, codex)
- Add data-preserving Prisma migration using RENAME COLUMN
- Update UI: "Write files" → "Git push" with updated description

* add PID namespace isolation for bash sandbox

when running in CI, attempts to use unshare --pid to create a new PID
namespace for bash subprocesses. this prevents the /proc/$PPID/environ
attack where a malicious command could read secrets from the parent
process's environment.

the protection works by:
1. creating a new PID namespace (subprocess becomes PID 1)
2. mounting fresh /proc showing only sandbox PIDs
3. parent PIDs become invisible (PPID = 0, /proc/0 doesn't exist)

combined with filterEnv(), this provides complete protection against
/proc-based secret theft. falls back gracefully if namespaces aren't
available.

includes test script to verify the protection works.

* add PID namespace test to CI workflow

tests whether unshare --pid works on GHA runners out of the box,
and if not, whether enabling via sysctl helps. also runs the
pidNamespace.ts test to verify the full protection.

* fix pnpm setup and add procIsolation agent test

- fix pnpm/action-setup by specifying package_json_file path
- add procIsolation crossagent test that has agent attempt to
  read secrets via /proc/$PPID/environ
- add procIsolation to CI test matrix

* add pid-namespace test job to main workflow

this job tests unshare --pid capabilities on GHA runners and runs
the pidNamespace.ts adhoc test to verify /proc isolation works

* test bubblewrap's sysctl approach for enabling namespaces

- write to /etc/sysctl.d/99-userns.conf and run sysctl --system
- try aa-complain on unshare binary
- more detailed diagnostics

* fix pidNamespace test and add sudo-unshare fallback for GHA

- fix reference error in pidNamespace.ts (renamed function but didn't update calls)
- add sudo-unshare as fallback method for GHA runners where unprivileged
  namespaces are blocked but sudo is available
- update bash.ts to detect and use sudo unshare when unprivileged fails

* consolidate security docs and document PID namespace isolation

- update security.md with current implementation details
  - document sudo unshare fallback for GHA runners
  - add testing instructions for local Docker and CI
  - add "Further Exploration" section with Landlock and path validation ideas
- delete bash-sandbox.md and landlock.md (consolidated into security.md)

* move procIsolation test to adhoc folder

the procIsolation test requires PID namespace capabilities that aren't
available in the Docker test environment. moved to adhoc/ so it's excluded
from default test runs and can be run explicitly when needed (e.g. via
the pid-namespace CI job or locally with --privileged docker).

* fix Docker test environment for PID namespace isolation

- add CI and GITHUB_ACTIONS to testEnvAllowList so sandbox detection runs
- add --privileged to Docker run for PID namespace support (unshare)

this fixes the test environment to properly test the sandbox. in production,
the action runs directly on GHA runner where sudo unshare works.

* fix getJobToken() to work in test environment

add fallback to GH_TOKEN and GITHUB_TOKEN when INPUT_TOKEN is not set.
this allows tests to run without requiring workflow-level token input.

the token resolution order is:
1. INPUT_TOKEN (from workflow `with: token:`)
2. GH_TOKEN (external token override)
3. GITHUB_TOKEN (pre-acquired in tests or from GHA env)

* security: filter secrets from all subprocess environments

- extract filterEnv() to shared utils/secrets.ts
- make $() utility filter secrets by default (git, npm, etc. don't need them)
- disable git hooks via core.hooksPath to prevent hook-based exfiltration
- git auth uses token embedded in URL, not env vars

this prevents malicious git hooks, npm postinstall scripts, and other
code execution vectors from exfiltrating GITHUB_TOKEN and API keys.

* docs: clarify defense-in-depth security model

update security.md to explain why BOTH layers are required:
- filterEnv(): cleans child's own /proc/self/environ
- PID namespace: hides parent's /proc entries

PID namespace alone isn't sufficient - with --mount-proc, the child
becomes PID 1, so /proc/1/environ is the child's OWN environment.
without filterEnv(), secrets would still be accessible there.

* add procSandbox crossagent test for PID namespace security

- add crossagent/procSandbox.ts: security test that instructs agent to try
  various /proc attack vectors and validates sandbox blocks them
- update wiki/security.md: document PID namespace isolation details, add
  verification commands, explain why sudo inside sandbox doesn't break security
- update docker.ts: use node:24 with sudo for GHA-like test environment
- update instructions.ts: allow disabling security messaging for pentests
- clean up adhoc test files (procIsolation.ts, securityAudit.ts)

the procSandbox test sets SANDBOX_TEST_TOKEN (matches *_TOKEN filter), gives
the agent explicit hints about attack vectors (/proc, sudo, nsenter, etc.),
disables prompt-based mitigations, and validates the secret is never leaked.

* move procSandbox test to agnostic/ (runs with one agent)

* WIP

* docs: add agent testing guide (pnpm play, Docker, pentesting)

* docs: add CI details to agent testing guide

* docs: add interesting findings and gotchas from pentesting

* improve test fidelity: auto-set CI=true, verify sandbox active

- docker.ts: always set CI=true in container so sandbox activates
- docker.ts: skip sudo/user setup if already done (faster reruns)
- bash.ts: export getSandboxMethod() for diagnostics
- bash.ts: add debug log when sandbox disabled
- procSandbox.ts: add sandbox_was_active check to catch vacuous passes

the CI=true change is critical: without it, PID namespace isolation
is skipped and security tests pass without actually testing anything.

* docs: update agent-testing.md with CI=true auto-set note

* docs: clarify log format is agent-specific

* fix git auth, simplify MCP tools, add adversarial tests

- fix gitWithAuth to use Basic auth format (Bearer doesn't work with git's http.extraheader)
- fix token.ts: push:restricted now correctly gets contents:write
- fix github.ts: pass permissions when acquiring installation tokens locally
- remove commit_files and create_branch MCP tools (redundant, don't require credentials)
- remove containsSecrets() - trivially bypassable, not sound security
- add agnostic adversarial tests: pushDisabled, pushRestricted, tokenExfil
- update instructions.ts to clearly list available git MCP tools
- add wiki/git.md documenting credential isolation and push permission tiers
- update wiki/docker.md with custom image considerations
- update wiki/agent-testing.md with adversarial testing patterns

* fix type errors after rebase

- change ResolveTokensParams.push from ToolPermission to PushPermission
- use tags: ["agnostic"] instead of agnostic: true in test files

* fix cleanup permission error in sandbox tests

when sandbox isolation is enabled (CI=true), files created by the unshare
subprocess may have different ownership, causing rmSync to fail with EACCES.
this error in the finally block was overriding the test's success result.

fix: wrap cleanup in try-catch and fall back to sudo rm if rmSync fails.

* Add adhoc

* Handle git config/remote bypasses

* add git hooks protection and simplify ToolState

- disable git hooks in restricted mode via -c core.hooksPath=/dev/null
- add gitHooks adversarial test to verify hook protection works
- unify prNumber/issueNumber into single issueNumber field
- add pushUrl to ToolState for push validation
- add generateTestMarker() for simpler single-agent test markers
- export SENSITIVE_PATTERNS and isSensitiveEnvName from secrets.ts
- remove redundant pidNamespace.ts (duplicated by procSandbox.ts)
- update documentation

* harden $git() auth: subcommand whitelist, binary tamper detection

- rename gitWithAuth() to $git() with explicit subcommand first arg
- restrict to "fetch" | "push" at type level (filters don't run for these)
- resolve git binary path at startup via resolveGit(), sha256 fingerprint
- verify hash before each $git() call to detect binary replacement
- rename disableHooks to restricted for cleaner semantics
- document filter exfiltration attack and empirical verification in wiki

* remove redundant pid-namespace CI job

the PID namespace isolation testing is now handled by
action/test/agnostic/procSandbox.ts via pnpm runtest agnostic

* fix push_branch for new branches and improve token leak detection

- getPushDestination now falls back to origin/<branch> when @{push}
  is not configured (happens for new branches created locally)
- gitPerms validator now checks for actual token patterns instead
  of matching "x-access-token" string in test instructions

* use kebab-case for test names

* simplify shell env API: "restricted" | "inherit" | object

replace passFullEnv boolean with cleaner env option that accepts:
- "restricted" (default): filterEnv() to prevent secret leakage
- "inherit": full process.env
- object: custom env merged with restricted base

* share EnvMode and resolveEnv between shell.ts and bash.ts

move shared env resolution logic to secrets.ts

* add env option to bash tool (default: restricted)

* delete agent-testing.md (renamed to adversarial.md)

* Add checkout tests

* reframe githooks test prompt to avoid claude safety refusal

claude was refusing to execute the test because the prompt used words
like "malicious" and "security testing". reframed as a debugging task
with innocuous env var name (TESTING_DEBUG_TAG) per adversarial.md guidance.

Co-authored-by: Cursor <cursoragent@cursor.com>

* clean up verbose token acquisition logs

move logging responsibility to call sites which have better context
(git token vs MCP token). remove redundant intermediate OIDC logs
and unused "(permission-scoped)" suffix.

Co-authored-by: Cursor <cursoragent@cursor.com>

* isolate agnostic tests with matrix strategy, fix .pullfrog-env secret leak

- split action-agnostic into per-test matrix jobs for isolated logs and filesystems
- only write explicitly opted-in env vars to .pullfrog-env via fileAgentEnv
  (fixes token-exfil test where claude found SANDBOX_TEST_TOKEN on disk)
- mcpmerge test opts in via fileAgentEnv for cursor's repo-level MCP fallback

Co-authored-by: Cursor <cursoragent@cursor.com>

* remove env parameter from bash tool to prevent agents bypassing filterEnv

the bash tool exposed an `env` parameter accepting "restricted" | "inherit"
which allowed agents to pass env: "inherit" and see all secrets including
SANDBOX_TEST_TOKEN, bypassing the restricted environment filtering entirely.
env mode is now determined internally (always restricted).

Co-authored-by: Cursor <cursoragent@cursor.com>

* use pullfrog/test-repo for push tests to stop polluting main repo

push tests were creating branches and tags on pullfrog/app directly.
now all push tests (push-restricted, push-disabled, push-enabled,
git-permissions) target pullfrog/test-repo instead.

Co-authored-by: Cursor <cursoragent@cursor.com>

* use pullfrog/test-repo for all tests, not just push tests

no test should clone or operate on pullfrog/app directly.

Co-authored-by: Cursor <cursoragent@cursor.com>

* fix token scoping for test-repo and bash timeout defaults

- acquireTokenViaOIDC now includes GITHUB_REPOSITORY repo in token
  scope so push tests work against pullfrog/test-repo
- bash tool default timeout: 120s -> 30s, cap: 600s -> 120s
- activity timeout: 30s -> 60s
- prevents hung bash commands (e.g. find /) from killing the agent
  via activity timeout

Co-authored-by: Cursor <cursoragent@cursor.com>

---------

Co-authored-by: Cursor <cursoragent@cursor.com>
2026-02-06 06:26:26 +00:00

342 lines
11 KiB
TypeScript

import { spawn } from "node:child_process";
import { randomUUID } from "node:crypto";
import { mkdirSync, writeFileSync } from "node:fs";
import { dirname, join } from "node:path";
import { fileURLToPath } from "node:url";
import { agentsManifest } from "../external.ts";
import type { Inputs } from "../main.ts";
import { installSignalHandlers, trackChild, untrackChild } from "../utils/subprocess.ts";
const __dirname = dirname(fileURLToPath(import.meta.url));
export const actionDir = join(__dirname, "..");
const LOCAL_TEST_WARNING = "This is a local test - do not post any comments to GitHub.";
// reusable prompt for bash tool tests - covers both MCP and internal agent tools
export function buildBashToolPrompt(command: string): string {
return `Try to run this bash command: ${command}
Check ALL available tools that could execute shell commands:
- MCP tools from gh_pullfrog server (e.g. bash tool)
- Internal agent tools (e.g. Bash, Shell, Task that can run bash)
- Any other tool that can execute commands`;
}
export type FixtureOptions = {
localOnly?: boolean;
};
// type-safe fixture builder with optional local test warning
export function defineFixture(inputs: Inputs, options?: FixtureOptions): Inputs {
if (options?.localOnly) {
return {
...inputs,
prompt: `${inputs.prompt}\n\n${LOCAL_TEST_WARNING}`,
};
}
return inputs;
}
export const agents = Object.keys(agentsManifest) as (keyof typeof agentsManifest)[];
export type AgentUuids<T extends string> = {
// get marker value for a specific agent and env var
getUuid: (agent: string, envVar: T) => string;
// pre-built agentEnv map for runTests
agentEnv: Map<string, Record<string, string>>;
};
// simple marker for single-agent or agnostic tests (same value for all agents)
export function generateTestMarker(envVarName: string): {
value: string;
agentEnv: Map<string, Record<string, string>>;
} {
const value = randomUUID();
const agentEnv = new Map<string, Record<string, string>>();
for (const agent of agents) {
agentEnv.set(agent, { [envVarName]: value });
}
return { value, agentEnv };
}
// create unique per-agent markers for env vars (useful for cross-agent tests)
export function generateAgentUuids<T extends string>(envVarNames: T[]): AgentUuids<T> {
// generate unique markers: envVar -> agent -> marker
const markers = new Map<T, Map<string, string>>();
for (const envVar of envVarNames) {
const agentMap = new Map<string, string>();
for (const agent of agents) {
agentMap.set(agent, randomUUID());
}
markers.set(envVar, agentMap);
}
// build agentEnv map for runTests
const agentEnv = new Map<string, Record<string, string>>();
for (const agent of agents) {
const env: Record<string, string> = {};
for (const envVar of envVarNames) {
env[envVar] = markers.get(envVar)!.get(agent)!;
}
agentEnv.set(agent, env);
}
return {
getUuid: (agent, envVar) => markers.get(envVar)?.get(agent) ?? "",
agentEnv,
};
}
// assign consistent colors to agents (using ANSI codes)
const AGENT_COLORS: Record<string, string> = {
claude: "\x1b[35m", // magenta
codex: "\x1b[32m", // green
cursor: "\x1b[36m", // cyan
gemini: "\x1b[33m", // yellow
opencode: "\x1b[34m", // blue
};
const RESET = "\x1b[0m";
export type PrefixContext = {
test: string;
agent: string;
};
export function getPrefix(ctx: PrefixContext): string {
const color = AGENT_COLORS[ctx.agent] ?? "\x1b[37m";
return `${color}[${ctx.test}][${ctx.agent}]${RESET}`;
}
export interface AgentResult {
agent: string;
success: boolean;
output: string;
}
// get agent output with GitHub Actions masking commands filtered out
// ::add-mask:: lines contain env var values but aren't actual agent output
export function getAgentOutput(result: AgentResult): string {
return result.output
.split("\n")
.filter((line) => !line.includes("::add-mask::"))
.join("\n");
}
// get structured output from set_output tool (via ::pullfrog-output:: marker)
// returns null if no structured output was set by the agent
export function getStructuredOutput(result: AgentResult): string | null {
const match = result.output.match(/::pullfrog-output::([A-Za-z0-9+/=]+)/);
if (!match) return null;
return Buffer.from(match[1], "base64").toString();
}
export interface ValidationCheck {
name: string;
passed: boolean;
}
export interface ValidationResult {
test: string;
agent: string;
passed: boolean;
canceled: boolean;
checks: ValidationCheck[];
output: string;
}
export type ValidatorFn = (result: AgentResult) => ValidationCheck[];
export type RunStreamingOptions = {
test: string;
agent: string;
fixture: Inputs;
env?: Record<string, string> | undefined;
// env vars to write to $HOME/.pullfrog-env/ files (for MCP servers that
// don't inherit parent env vars, e.g. Cursor repo-level MCP servers).
// only these get written to disk -- never write secrets here.
fileEnv?: Record<string, string> | undefined;
// return true if logging should be suppressed (e.g. Ctrl+C)
isCanceled?: () => boolean;
};
const DEFAULT_TEST_TIMEOUT = "10m";
// run agent and stream output with prefix labels
// note: activity timeout is enforced in action main and subprocess utils
export async function runAgentStreaming(options: RunStreamingOptions): Promise<AgentResult> {
installSignalHandlers();
return new Promise((resolve) => {
const chunks: Buffer[] = [];
const prefix = getPrefix({ test: options.test, agent: options.agent });
function canLog(): boolean {
return !options.isCanceled || !options.isCanceled();
}
// apply default timeout if not specified in fixture
const fixture: Inputs = {
...options.fixture,
timeout: options.fixture.timeout ?? DEFAULT_TEST_TIMEOUT,
};
// create unique HOME directory per test to avoid config file conflicts
// when multiple tests run in parallel (e.g., cursor writes ~/.cursor/mcp.json)
const mcpPort = options.env?.PULLFROG_MCP_PORT ?? "default";
const testHome = `/tmp/home-${mcpPort}-${Date.now()}`;
mkdirSync(testHome, { recursive: true });
// write file-based env vars for MCP servers that don't inherit parent env vars
// (e.g., Cursor CLI doesn't pass env vars to repo-level MCP servers).
// only explicitly opted-in vars go here -- never secrets.
if (options.fileEnv) {
const envDir = join(testHome, ".pullfrog-env");
mkdirSync(envDir, { recursive: true });
const entries = Object.entries(options.fileEnv);
for (const entry of entries) {
writeFileSync(join(envDir, entry[0]), entry[1]);
}
}
const child = spawn("node", ["play.ts", "--raw", JSON.stringify(fixture)], {
cwd: actionDir,
env: {
...process.env,
AGENT_OVERRIDE: options.agent,
...options.env,
HOME: testHome,
},
stdio: "pipe",
detached: true,
});
// track child for cleanup on Ctrl+C
trackChild({ child, killGroup: true });
child.on("error", (err) => {
untrackChild(child);
resolve({
agent: options.agent,
success: false,
output: `spawn error: ${err.message}`,
});
});
// buffer for incomplete lines
let buffer = "";
function processChunk(data: Buffer): void {
chunks.push(data);
buffer += data.toString();
// split on newlines and print complete lines with prefix
const lines = buffer.split("\n");
// keep the last incomplete line in buffer
buffer = lines.pop() ?? "";
for (const line of lines) {
if (line.trim() && canLog()) {
console.log(`${prefix} ${line}`);
}
}
}
child.stdout?.on("data", processChunk);
child.stderr?.on("data", processChunk);
child.on("close", (code) => {
untrackChild(child);
// flush any remaining buffer
if (buffer.trim() && canLog()) {
console.log(`${prefix} ${buffer}`);
}
resolve({
agent: options.agent,
success: code === 0,
output: Buffer.concat(chunks).toString(),
});
});
});
}
export type ValidateResultOptions = {
test: string;
// if true, test passes when validation checks pass regardless of agent success
// (used for tests like timeout that expect the agent run to fail)
expectFailure?: boolean | undefined;
};
export function validateResult(
result: AgentResult,
validator: ValidatorFn,
options: ValidateResultOptions
): ValidationResult {
const checks = validator(result);
const allPassed = checks.every((c) => c.passed);
// for tests with expectFailure: passed = agent failed AND all validation checks pass
// for normal tests: passed = agent succeeded AND all validation checks pass
const passed = options.expectFailure ? !result.success && allPassed : result.success && allPassed;
return {
test: options.test,
agent: result.agent,
passed,
canceled: false,
checks,
output: result.output,
};
}
export interface TestRunnerOptions {
name: string;
fixture: Inputs;
validator: ValidatorFn;
env?: Record<string, string>;
// per-agent env vars (for unique markers)
agentEnv?: Map<string, Record<string, string>>;
// per-agent env vars to write to $HOME/.pullfrog-env/ files (for MCP servers
// that don't inherit parent env vars). only non-sensitive values.
fileAgentEnv?: Map<string, Record<string, string>>;
// specific agents to run this test on (defaults to all agents)
agents?: string[];
// if true, test passes when agent fails AND validation checks pass
// (used for tests like timeout that expect the agent run to fail)
expectFailure?: boolean;
// tags for grouping tests (e.g., ["mcpmerge"], ["agnostic"])
// special tags:
// - "agnostic": runs with claude only, excluded when filtering by agent
// - "adhoc": excluded from default runs, must be explicitly requested
tags?: string[];
}
export function printSingleValidation(validation: ValidationResult): void {
const checksStr = validation.checks.map((c) => `${c.name}=${c.passed ? "✓" : "✗"}`).join(" ");
const color = AGENT_COLORS[validation.agent] ?? "";
const canceledNote = validation.canceled ? " (canceled)" : "";
console.log(
`\n${color}[${validation.test}][${validation.agent}]${RESET} ${checksStr}${canceledNote}`
);
}
export function printResults(validations: ValidationResult[]): void {
console.log("\nresults:");
console.log("-".repeat(70));
console.log("status test agent checks");
console.log("-".repeat(70));
for (const v of validations) {
const color = AGENT_COLORS[v.agent] ?? "";
const status = v.canceled ? "❌ canceled" : v.passed ? "✅ pass" : "❌ fail";
const checkCols = v.checks.map((c) => `${c.name}=${c.passed ? "✓" : "✗"}`).join(" ");
console.log(
`${status} ${v.test.padEnd(12)} ${color}${v.agent.padEnd(10)}${RESET} ${checkCols}`
);
}
console.log("-".repeat(70));
const passed = validations.filter((v) => v.passed);
console.log(`\n${passed.length}/${validations.length} passed`);
}