greydgl--pentestgpt
ab5fbb4d90
* first refactor
* feat: dockerized tool with persistent Claude+Codex login + multi-model benchmark
Run the autonomous CTF/pentest tool in Docker with a one-time, persistent login for
BOTH Claude Code and Codex, and add a multi-model benchmark harness.
Backend (multi-model):
- Add `--backend {claude,codex}` to the CTF pipeline. CodexBackend (pentestgpt/core/
backend.py) wraps unified_agent's Codex backend and translates its events into
AgentMessages, so the same pipeline runs on Claude (opus/sonnet) or Codex
(gpt-5.5/gpt-5.4-mini). Wired through config.backend, pipeline stage construction,
and the CLI (+ PENTESTGPT_CODEX_EFFORT; greppable [CODEX_USAGE] under PENTESTGPT_BENCH=1).
Docker tool (tool-only image; the benchmark stays OUTSIDE the image):
- Extend Dockerfile: Codex CLI (@openai/codex) + openai_codex SDK + unified_agent/
pentestgpt_agent/pentestgpt_legacy packages + gobuster/dirb + socat. Add .dockerignore
(keeps creds/benchmark/workspace out of the build context).
- Persistent dual login (the hard part) — asymmetric by token model:
* Claude: `setup-token` -> token stored in the pentestgpt-claude volume; entrypoint
exports CLAUDE_CODE_OAUTH_TOKEN (setup-token does not write .credentials.json; macOS
host creds live in the Keychain and can't be copied).
* Codex: the container does its OWN `codex login` (NOT seeding -- ChatGPT refresh tokens
are single-use, so a shared/copied login 401s on first refresh). The 127.0.0.1:1455
OAuth callback is forwarded into the container via a socat hop (-p 1455:8455).
* scripts/docker-login.sh is idempotent: checks logins live, logs in only the missing one(s).
- docker-compose codex-config volume (+ pinned names); entrypoint token-export + non-blocking
preflight; scripts/docker-auth-status.sh; Make targets (docker-build/login/auth-status/
run/shell/down/nuke).
- Verified end-to-end: one `make docker-login` -> a fresh container reports claude+codex
logged in with live round-trips; the CTF pipeline (Codex) captured a flag against an
isolated fixture and the pentest pipeline ran cleanly; persists across recreation, no re-login.
Benchmark (multi-model, host-side):
- benchmark/pilot/ harness (run_pilot.py + report.py): builds each xbow challenge, discovers
the loopback port, runs the pipeline across the 4 model combos, judges by the baked
FLAG{sha256(UPPER-dir)}, and renders REPORT.md (infra failures excluded from solve rates).
Includes the partial pilot's results (results.jsonl + REPORT.md).
Docs: docs/docker-dev-plan.md (full plan + implementation status); CLAUDE.md and README
docker quickstart; benchmark/pilot/README.md; design-doc roadmap (docs/redesign).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* fix: fail controller on backend error messages
* fix: allow listing sessions without target
* docs: add docker xbow benchmark report
* fix: infer concrete backend constructor type
* docs: refresh docker benchmark documentation
* feat(benchmark): add pure single-agent baseline + pipeline comparison
Add a "pure single agent" benchmark variant -- one bare `claude -p` /
`codex exec` call per target (no pipeline) -- to quantify what the 3-stage
PentestGPT pipeline buys over an un-orchestrated agent on the xbow targets.
- pentestgpt/prompts/stages.py: ctf_single_agent_{system,task}_prompt -- the
pipeline's shared fragments collapsed into ONE turn, so prompt content is
held constant and the only variable is the multi-stage decomposition.
- benchmark/pilot/run_docker_bench.py: docker-network runner
(--variant single|pipeline). Brings the target up, discovers the container's
internal IP+network (skips DB side-cars/ports), docker-runs the tool image on
that network, and scores the ground-truth flag against the agent's *assistant
text* only (parity with the pipeline's raw streaming). Reads stdout in chunks
to handle >64KB JSON lines. Resumable; --dry-run supported.
- benchmark/pilot/report_comparison.py -> DOCKER_COMPARISON.md: head-to-head
pipeline-vs-single per model on the common non-infra set.
- tests/unit/test_single_agent_prompt.py: prompt-builder coverage.
- docs: README, CLAUDE.md, benchmark README, DOCKER_REPORT updated.
Recorded result (10 medium/hard targets x 4 models, container-to-container,
same baseline image digest 0c4c0f3e..., commit dca0019 image):
Model Pipeline Single
Claude Opus 5/10 7/10 (single +2)
Claude Sonnet 6/10 4/10 (pipeline +2)
Codex gpt-5.5 7/10 7/10 (tie)
Codex gpt-5.4-mini 3/10 4/10 (single +1)
TOTAL 21/40 22/40
Single agent matches the pipeline on solve rate (55% vs 52%) while using
~40% fewer Codex tokens (13.0M vs 21.8M) and solving faster. The pipeline
only clearly helps Claude Sonnet (which times out solo); Opus is better solo.
Full per-challenge grid in DOCKER_COMPARISON.md; raw records in
docker_single_results.jsonl.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* feat(benchmark): add pentestgpt_agent docker harness
* bench: refresh pentestgpt_agent smoke result
* fix(benchmark): make repeat rows variant-aware
* fix(agent): fall back for semantic executor labels
* fix(agent): tolerate executor prose evidence
* fix(benchmark): score accepted framework findings
* bench: append partial framework repeat results
* bench: complete framework repeat sweep
* bench: expose framework executor concurrency
* bench: add extended parallel framework sweep
* checkpoint: preserve working agent and benchmark state
* feat: harden durable agent loop and xbow qualification
* fix: reserve an exploit result turn
* docs: record clean xbow qualification
* build: consume unified-agent from the git wrapper repo
Repoint pentestgpt_agent_new's unified-agent dependency from the local
editable path (../../UnifiedAgentPoC, now renamed and gone) to the pinned
git source PentestGPT-Project/UnifedAgentWrapper@d05d21f. Regenerate uv.lock
and update test_dependency.py to assert the external package is installed
from that VCS URL (not the repo-root vendored copy) at version 0.2.0.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* refactor: make pentestgpt_agent_new the sole framework
Remove the retired ledger-based pentestgpt_agent package (instructor/executor/
judge) and its orphaned unit + smoke tests. The nested pentestgpt_agent_new
project (Supervisor/Executor over a durable SQLite loop, consuming unified-agent
from the git wrapper) is now the single maintained framework.
Repoint the top-level tooling to it:
- pyproject: drop the pentestgpt-agent console script and pentestgpt_agent from
the wheel packages.
- Makefile: lint/format target parent code only; typecheck/check/ci now run the
nested framework's own gate (ruff, format, mypy, pytest) via test-agent-new /
check-agent-new, so `make check` finally covers it; `make run` delegates to the
pentestgpt-agent-new CLI.
- Dockerfile: stop copying the removed package (kept the build working); note the
framework is not baked into the image yet.
- docker container-health test: import the substrate packages that actually ship.
- CLAUDE.md / AGENT.md: describe the new framework, the git-sourced wrapper, and
the deprioritized benchmark/Docker rewire.
The XBOW `--variant framework` path and docker-bench Makefile targets still point
at the old in-image framework and are left as a pending rewire (benchmarks
deprioritized); the naive `--variant single` path is unaffected.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* refactor: rename pentestgpt_agent_new -> pentestgpt_agent
The framework reclaims the clean name now that the old ledger-based package is
gone. Rename the nested project folder, its src package, the distribution
(pentestgpt-agent-new -> pentestgpt-agent) and CLI, and every import/reference in
the package, the umbrella Makefile, the Dockerfile, the docker health test, and
CLAUDE.md / AGENT.md. Regenerate uv.lock. The audit CLI stays pentestgpt-agent-audit;
the git-sourced unified-agent dependency is unchanged. `make check` is green
(108 nested tests). The two historical *_REPORT.md files keep the old name as
dated records.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* chore: extract benchmark harness to sibling xbow-benchmark repo
Move PentestGPT/benchmark/ out to ../xbow-benchmark (its own repo) to keep this
project clean. The harness was decoupled from the framework code (it scores
container output, never imports pentestgpt_agent/unified_agent), so only
operational ties remain and they now live in the sibling repo.
- Remove benchmark/ and the 4 harness unit tests (relocated + repointed there).
- Strip the docker-bench-*/bench-* targets and their config vars from the
Makefile; keep the tool-image lifecycle (docker-build/login/run/...) and add a
help pointer to `make -C ../xbow-benchmark help`.
The sibling repo mounts this checkout read-only (--source-root ../PentestGPT) and
runs the pentestgpt:latest image built here.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* feat: harden autonomous framework and runtime integration
---------
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
188 行
6.1 KiB
Python
188 行
6.1 KiB
Python
import asyncio
|
|
import time
|
|
|
|
import pytest
|
|
|
|
from unified_agent.agent import SuperAgent, UnifiedAgent
|
|
from unified_agent.events import (
|
|
AssistantText,
|
|
FileChanged,
|
|
SessionStarted,
|
|
ToolCall,
|
|
TurnCompleted,
|
|
)
|
|
from unified_agent.types import BackendUnavailableError, UnifiedUsage
|
|
|
|
|
|
class FakeBackend:
|
|
name = "fake"
|
|
|
|
def __init__(self, events=None, delay=0.0, fail_with=None, fake_name="fake"):
|
|
self.name = fake_name
|
|
self.events = events or []
|
|
self.delay = delay
|
|
self.fail_with = fail_with
|
|
self.seen_prompts = []
|
|
self.seen_opts = []
|
|
|
|
async def stream(self, prompt, opts):
|
|
self.seen_prompts.append(prompt)
|
|
self.seen_opts.append(opts)
|
|
if self.delay:
|
|
await asyncio.sleep(self.delay)
|
|
for ev in self.events:
|
|
yield ev
|
|
if self.fail_with:
|
|
raise self.fail_with
|
|
|
|
|
|
GOOD_EVENTS = [
|
|
SessionStarted(session_id="s-1"),
|
|
AssistantText("working on it"),
|
|
ToolCall(name="mcp__unified__add_numbers", input={"a": 1, "b": 2}, call_id="c1"),
|
|
FileChanged(path="out.txt", kind="add"),
|
|
TurnCompleted(
|
|
success=True,
|
|
final_text="all done",
|
|
usage=UnifiedUsage(input_tokens=10, output_tokens=5),
|
|
cost_usd=0.01,
|
|
session_id="s-1",
|
|
duration_ms=123,
|
|
),
|
|
]
|
|
|
|
|
|
def make_agent(tmp_path, backend) -> UnifiedAgent:
|
|
return UnifiedAgent(backend=backend, workspace=tmp_path / "ws")
|
|
|
|
|
|
async def test_run_collects_unified_result(tmp_path):
|
|
backend = FakeBackend(events=GOOD_EVENTS)
|
|
agent = make_agent(tmp_path, backend)
|
|
result = await agent.run("do the thing")
|
|
assert result.success is True
|
|
assert result.backend == "fake"
|
|
assert result.text == "all done"
|
|
assert result.usage.input_tokens == 10
|
|
assert result.cost_usd == 0.01
|
|
assert result.session_id == "s-1"
|
|
assert [t.name for t in result.tool_calls] == ["mcp__unified__add_numbers"]
|
|
assert [f.path for f in result.file_changes] == ["out.txt"]
|
|
assert len(result.events) == len(GOOD_EVENTS)
|
|
assert result.error is None
|
|
|
|
|
|
async def test_run_text_falls_back_to_last_assistant_text(tmp_path):
|
|
events = [AssistantText("a"), AssistantText("b"), TurnCompleted(success=True)]
|
|
result = await make_agent(tmp_path, FakeBackend(events=events)).run("x")
|
|
assert result.text == "b"
|
|
|
|
|
|
async def test_prompt_rendered_for_backend_name(tmp_path):
|
|
backend = FakeBackend(events=GOOD_EVENTS, fake_name="codex")
|
|
agent = make_agent(tmp_path, backend)
|
|
from unified_agent.task import Task
|
|
|
|
await agent.run(Task(instruction="go", skill="my-skill"))
|
|
assert "$my-skill" in backend.seen_prompts[0]
|
|
|
|
|
|
async def test_run_failure_returns_result_not_exception(tmp_path):
|
|
backend = FakeBackend(events=[AssistantText("partial")], fail_with=RuntimeError("boom"))
|
|
result = await make_agent(tmp_path, backend).run("x")
|
|
assert result.success is False
|
|
assert "boom" in result.error
|
|
assert result.text == "partial" # falls back to last assistant text
|
|
|
|
|
|
async def test_stream_propagates_exception(tmp_path):
|
|
backend = FakeBackend(fail_with=RuntimeError("boom"))
|
|
agent = make_agent(tmp_path, backend)
|
|
with pytest.raises(RuntimeError):
|
|
async for _ in agent.stream("x"):
|
|
pass
|
|
|
|
|
|
async def test_run_options_carry_configuration(tmp_path):
|
|
backend = FakeBackend(events=GOOD_EVENTS)
|
|
agent = UnifiedAgent(
|
|
backend=backend,
|
|
workspace=tmp_path / "ws",
|
|
model="some-model",
|
|
instructions="be terse",
|
|
effort="xhigh",
|
|
)
|
|
schema = {"type": "object", "properties": {"x": {"type": "string"}}}
|
|
await agent.run("t", output_schema=schema, resume="sess-9", max_turns=3)
|
|
opts = backend.seen_opts[0]
|
|
assert opts.model == "some-model"
|
|
assert opts.instructions == "be terse"
|
|
assert opts.effort == "xhigh"
|
|
assert opts.output_schema == schema
|
|
assert opts.resume == "sess-9"
|
|
assert opts.max_turns == 3
|
|
assert opts.workspace == (tmp_path / "ws").resolve()
|
|
assert (tmp_path / "ws").is_dir() # created by prepare
|
|
|
|
|
|
async def test_tools_and_skills_prepared_once(tmp_path):
|
|
from unified_agent.skills import make_skill
|
|
|
|
skills_src = tmp_path / "skills"
|
|
make_skill(skills_src, "alpha-skill", "Does alpha.", "Body")
|
|
backend = FakeBackend(events=GOOD_EVENTS)
|
|
agent = UnifiedAgent(
|
|
backend=backend,
|
|
workspace=tmp_path / "ws",
|
|
tools="tests.fixture_registry:REG",
|
|
skills_dir=skills_src,
|
|
)
|
|
await agent.run("a")
|
|
await agent.run("b")
|
|
opts = backend.seen_opts[0]
|
|
assert opts.tool_server is not None
|
|
assert opts.tool_server.server_name == "unified"
|
|
assert (tmp_path / "ws" / ".claude" / "skills" / "alpha-skill").exists()
|
|
assert (tmp_path / "ws" / ".agents" / "skills" / "alpha-skill").exists()
|
|
|
|
|
|
def test_unknown_backend_name_raises():
|
|
with pytest.raises(BackendUnavailableError):
|
|
UnifiedAgent(backend="gemini")
|
|
|
|
|
|
def test_run_sync(tmp_path):
|
|
backend = FakeBackend(events=GOOD_EVENTS)
|
|
result = make_agent(tmp_path, backend).run_sync("x")
|
|
assert result.success
|
|
|
|
|
|
async def test_superagent_runs_all_concurrently(tmp_path):
|
|
a = UnifiedAgent(
|
|
backend=FakeBackend(events=GOOD_EVENTS, delay=0.2, fake_name="a"), workspace=tmp_path / "wa"
|
|
)
|
|
b = UnifiedAgent(
|
|
backend=FakeBackend(events=GOOD_EVENTS, delay=0.2, fake_name="b"), workspace=tmp_path / "wb"
|
|
)
|
|
squad = SuperAgent({"a": a, "b": b})
|
|
t0 = time.monotonic()
|
|
results = await squad.run_all("same task")
|
|
elapsed = time.monotonic() - t0
|
|
assert set(results) == {"a", "b"}
|
|
assert all(r.success for r in results.values())
|
|
assert elapsed < 0.35, f"not concurrent: {elapsed:.2f}s"
|
|
|
|
|
|
async def test_superagent_run_one_and_subset(tmp_path):
|
|
a = UnifiedAgent(
|
|
backend=FakeBackend(events=GOOD_EVENTS, fake_name="a"), workspace=tmp_path / "wa"
|
|
)
|
|
b = UnifiedAgent(
|
|
backend=FakeBackend(events=GOOD_EVENTS, fake_name="b"), workspace=tmp_path / "wb"
|
|
)
|
|
squad = SuperAgent({"a": a, "b": b})
|
|
r = await squad.run("a", "task")
|
|
assert r.backend == "a"
|
|
only_b = await squad.run_all("task", only=["b"])
|
|
assert set(only_b) == {"b"}
|