Line 72 was over-indented, causing an IndentationError on import of strix/report/writer.py and breaking main. Also bump the mirrors-mypy pre-commit hook to v1.17.1 to avoid the mypy 1.16.0 internal crash (python/mypy#19412) on openai/_client.py.
* fix: resolve pre-commit check failures
- Change RuntimeError to TypeError for type validation in report/writer.py
- Update pyupgrade to v3.21.2 for Python 3.14 compatibility
* chore: add pytest test infrastructure
Mirror the layout introduced on feature/438-token_budget: pytest +
pytest-asyncio dev deps, asyncio_mode auto, a tests.* mypy override, and
pytest in the mypy pre-commit hook deps so the tests/ package type-checks.
* feat: add --mount and large-target pre-flight for local repos (#492)
Large local targets were copied into the sandbox file-by-file via the SDK
LocalDir entry, which stalls on big repos and could leave /workspace empty.
- --mount <path> bind-mounts a host directory read-only at /workspace/<subdir>
instead of copying it, bypassing the per-file stream.
- A size pre-flight (STRIX_MAX_LOCAL_COPY_MB, default 1024) fails fast with a
clear message suggesting --mount when a non-mounted local target is too big.
* fix: reject empty --mount paths
An empty or whitespace-only --mount value resolves to the current working
directory and would silently bind-mount it into the sandbox. Reject it.
* fix: dedupe local targets so a dir is never both copied and mounted
If the same directory is passed via --target and --mount (or as duplicate
values), it previously produced two targets — copied AND bind-mounted, and
the copied one could trip the size pre-flight. Dedupe by resolved path,
preferring the bind mount.
* fix: treat non-positive STRIX_MAX_LOCAL_COPY_MB as disabled
Previously a value of 0 (or negative) made every local target count as
oversized, aborting all local scans. Now <= 0 disables the pre-flight.
* fix: log unreadable subtrees during size pre-flight
os.walk silently swallowed directory-listing errors, so a permission-denied
subtree could make a large repo under-count and slip past the pre-flight.
Surface such omissions via an onerror warning.
* docs: document --mount and STRIX_MAX_LOCAL_COPY_MB
Add CLI reference + example for --mount, document the size pre-flight env var,
note the read-only-is-not-a-hard-boundary caveat and that remote repos are not
size-checked, and clarify the backends docstring on when bind mounts apply.
* Update strix/interface/main.py
* Update strix/runtime/docker_client.py
---------
* fix: resolve pre-commit check failures
- Change RuntimeError to TypeError for type validation in report/writer.py
- Update pyupgrade to v3.21.2 for Python 3.14 compatibility
* feat(cli): add --max-budget-usd flag
Raises BudgetExceededError in ReportUsageHooks after each LLM call when
accumulated cost reaches the limit, with clean "stopped" status and
child-agent cancellation in non-interactive mode.
* test: add budget enforcement unit tests
7 tests covering no-budget, under-budget, at-limit, over-limit, error
message content, None report state, and exception hierarchy.
Also adds pytest/pytest-asyncio to dev deps and a mypy override for tests.
* fix(budget): validate positive budget and check the live cost ledger
Two hardening fixes for --max-budget-usd enforcement:
- Reject non-positive budgets. ReportUsageHooks now raises ValueError for
max_budget_usd <= 0, and the CLI validates the flag via a custom argparse
type so '--max-budget-usd 0' fails fast with a friendly message instead of
silently killing the scan on the first model response.
- Read the live cost. The budget check now reads ReportState.get_total_llm_cost()
(the live ledger) instead of the persisted run-record snapshot, so it stays
accurate even when a usage save fails after a model call.
* fix(budget): stop the entire scan deterministically when the limit is hit
Previously a BudgetExceededError was handled per-agent: it was swallowed in
interactive mode (the loop kept waiting), a child's error escaped its detached
task as an unretrieved-exception warning, the parent was never released from
wait_for_message, and the stop was logged at ERROR with a traceback as if the
agent had failed.
Replace that with a single scan-wide signal on the coordinator:
- AgentCoordinator.trigger_budget_stop() sets a flag and wakes every parked
agent; wait_for_message returns as soon as the flag is set.
- The run loops check coordinator.budget_stopped and raise to exit cleanly,
marking themselves 'stopped'. The root's exception reaches run_strix_scan's
handler, which cancels descendants and tears the scan down once; child
exceptions are swallowed in their detached task.
- The budget stop is logged at INFO, not as a failure.
This is deterministic regardless of tree depth or which agent first sees the
limit, fixing the interactive/TUI hang where a deep agent's stop never reached
a parked root. Also re-raises BudgetExceededError explicitly in the stream
handler so it can't be mistaken for the LiteLLM 'after shutdown' race.
* fix(budget): treat a budget stop as a clean stop in the TUI
Add an explicit BudgetExceededError handler in the TUI scan thread so that, if
the error ever reaches it, the budget stop is logged as a graceful stop rather
than surfaced as a red scan error by the broad 'except Exception'. The runner
normally absorbs the error and returns cleanly, so this is defensive depth for
a money-spending feature.
* docs(cli): document --max-budget-usd behavior and limitations
Clarify that the budget is cumulative across all agents, checked after each
model response, that the scan stops cleanly (not as a failure), that the value
must be > 0, and that spend can slightly overshoot due to in-flight calls and
best-effort cost estimation.
* Apply suggestions from code review
Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com>
---------
Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com>
Register a litellm.success_callback that captures kwargs['response_cost']
into a new observed-cost bucket on LLMUsageLedger. record() skips the
tokens-times-registry estimate for LiteLLM-routed models so we do not
double-count with the callback; OpenAI direct routes keep estimating
since LiteLLM is not invoked for them. Per-agent attribution for
LiteLLM-routed calls is apportioned by token share at to_record() time.
OpenAI's Responses API rejects reasoning.effort on non-reasoning
models like gpt-4o with `unsupported_parameter`, so any scan with
the default STRIX_REASONING_EFFORT=high against gpt-4o crashed at
the first model call. drop_params=True absorbs the rejected param
on LiteLLM-routed models but the SDK's native OpenAI path has no
equivalent.
Lift model_supports_reasoning to a public helper that strips
litellm/, any-llm/, openai/ prefixes and falls back to last-segment
lookup so prefixed forms like anthropic/claude-opus-4-7 resolve
through the bare model_cost entry. make_model_settings regains
model_name and skips Reasoning() when the registry doesn't confirm
support. uses_chat_completions_tool_schema reuses the same helper
(was duplicating the lookup under a misleading name).
OpenAI's Responses API rejects tools[i].type="custom" on non-reasoning
models like gpt-4o (400 with code=unknown_parameter, param=tools).
Strix's SDK-native Filesystem capability registers CustomTool entries
by default, so a bare STRIX_LLM=gpt-4o run failed at the first tool
invocation even though warm-up (a tool-less call) succeeded.
uses_chat_completions_tool_schema now consults
litellm.model_cost[<name>].supports_reasoning for OpenAI routes and
flips to the chat-completions function-tool schema for models that
don't carry the reasoning flag. Same registry-lookup pattern as
is_known_openai_bare_model. Non-OpenAI prefixes and configs with
LLM_API_BASE are unchanged (still function tools).
litellm.suppress_debug_info silences two unsolicited print() calls in
LiteLLM core: the "Provider List: https://docs.litellm.ai/docs/providers"
banner emitted by get_llm_provider_logic and the "Give Feedback /
Get Help" + "If you need to debug this error, use litellm._turn_on_debug()"
pair emitted by exception_mapping_utils on every LiteLLM exception.
Both are unconditional print() calls, not logger output, so log-level
config can't catch them. LiteLLM's own router and proxy_server set the
same flag for the same reason.
Warn on bare unknown model names before warm-up. is_known_openai_bare_model
consults litellm.model_cost and matches only entries whose
litellm_provider == "openai". When the configured STRIX_LLM has no
provider prefix, isn't a known OpenAI model, and no LLM_API_BASE is
set, show a clear panel pointing the user at the <provider>/<model>
form and exit before issuing the doomed request — no more chasing an
"Incorrect API key" 401 from OpenAI when the user actually meant
deepseek/, anthropic/, etc. Custom-base configs are still allowed
through unconfirmed.
Disable LiteLLM's message-logging and streaming-logging knobs to cut
noise and skip one of the two end-of-stream submit paths. The other
path at streaming_handler.py:2206 schedules work on a global
ThreadPoolExecutor that loses to atexit shutdown when the interpreter
is winding down; the SDK's stream consumer surfaces that as a fatal
"cannot schedule new futures after shutdown" RuntimeError even though
the actual stream content was already delivered. Catch and swallow
that specific RuntimeError in _run_cycle so the scan isn't killed by
an upstream end-of-stream logging race.
Drop every hand-rolled provider table and per-model gating that had
accumulated in the model-handling layer:
* normalize_model_name no longer auto-prefixes bare claude-* / gemini-*
names. Users supply the full <provider>/<model> form. The function
became literally model_name.strip(), so callers now inline that and
the function is removed.
* tool_choice="required" is gone everywhere. Thinking-mode endpoints
(Anthropic, DeepSeek /beta) reject it; modern reasoning models don't
need it; non-interactive runs already have
_append_noninteractive_tool_required_message as the convergence
backstop. model_supports_reasoning, model_known_to_registry, and
_model_cost_entry were only used to gate this and follow it out.
* Reasoning(effort=...) is now attached whenever
STRIX_REASONING_EFFORT is non-none. litellm.drop_params=True absorbs
it for non-reasoning models.
* Warm-up's bare-name OpenAI 401 hint is removed (false-positive prone,
relied on substring matching).
* reset_tool_choice on SandboxAgent is no-op now (no tool_choice gets
set) and is removed.
* report/dedupe.py was still routing through stock MultiProvider, so
non-OpenAI configs failed the dedupe LLM pass; switch it to
StrixProvider.
Verified end-to-end against modern provider strings (openai/gpt-5.4,
anthropic/claude-opus-4-7, deepseek/deepseek-reasoner,
gemini/gemini-2.5-pro, groq/, xai/, mistral/, together_ai/, perplexity/,
openrouter/, litellm/ legacy form, and whitespace-padded input): 18/18
cases route correctly, env vars mirror via litellm.validate_environment,
and ModelSettings carries no tool_choice. mypy strict passes.
Users had to type STRIX_LLM=litellm/deepseek/deepseek-chat — the
litellm/ wrapper was Strix-internal plumbing surfacing in user config.
Add StrixProvider, a MultiProvider subclass that routes any non-OpenAI
prefix (deepseek/, anthropic/, groq/, xai/, mistral/, openrouter/, …)
through LitellmProvider with the prefix preserved. normalize_model_name
no longer adds litellm/ to anything; bare claude-* / gemini-* shorthands
expand to anthropic/<model> / gemini/<model> instead of the wrapped form.
Wire StrixProvider into warm_up_llm and RunConfig.model_provider.
litellm/<provider>/<model> and any-llm/<provider>/<model> still resolve
unchanged for users on older config.
Refresh stale model names in the env-validation messages and the
warm-up hint (gpt-5.4, claude-opus-4-7, deepseek-reasoner).
Verified 24-case end-to-end matrix: OpenAI direct vs. LitellmProvider
routing, env-var mirroring via validate_environment, supports_reasoning
detection, and tool_choice gating all behave correctly across modern
providers including the user's unknown DeepSeek SKU.
When the user opts into reasoning_effort but the configured model
isn't in litellm.model_cost at all (private SKUs, fresh releases the
registry hasn't picked up — e.g. deepseek/deepseek-v4-pro), we can't
confirm thinking support and were sending tool_choice="required",
which thinking-mode endpoints reject ("Thinking mode does not support
this tool_choice").
Add model_known_to_registry() and split the decision: when the user
wants reasoning AND the model is either confirmed-reasoning OR
unknown-to-registry, drop tool_choice. The Reasoning(effort=...) param
still only attaches for confirmed-reasoning models, so we don't send
reasoning hints to known non-reasoning models.
Known non-reasoning models (gpt-4o, registry-confirmed) keep
tool_choice="required" unchanged.
A bare model name without a provider prefix routes through the SDK's
default OpenAI provider, so configuring STRIX_LLM=deepseek-v4-pro with
LLM_API_KEY=<deepseek key> sends that key to api.openai.com and
surfaces a confusing "Incorrect API key" error pointing at the OpenAI
dashboard.
When warm-up fails with an OpenAI-shaped error AND the configured
model is still unprefixed after normalize_model_name, append a hint
that points the user at the '<provider>/<model>' form with concrete
examples.
Naively uppercasing the routing prefix breaks for providers whose
LiteLLM env var name doesn't match the prefix verbatim:
together_ai/... needs TOGETHERAI_API_KEY (no underscore)
perplexity/... needs PERPLEXITYAI_API_KEY
Ask LiteLLM directly via litellm.validate_environment(model=...) which
env vars it consults for the chosen provider, then setdefault each one
to LLM_API_KEY. This is the SDK-blessed lookup and stays correct for
every provider LiteLLM supports without a hand-maintained name map.
Lowercase the routed model name before lookup so mixed-case user input
(e.g. Together_AI/...) still resolves.
normalize_model_name expands `claude-*` and `gemini-*` shorthands into
`litellm/anthropic/...` and `litellm/gemini/...` at routing time, but
the mirror helper was looking at the raw pre-normalization name — bare
shorthands had no `/` and hit the early return, so ANTHROPIC_API_KEY /
GEMINI_API_KEY were never populated for those users.
Run the same normalization inside the mirror helper so the provider
prefix is consistent with what LiteLLM actually sees downstream.
LiteLLM's per-provider branches (deepseek, anthropic, groq, etc.)
don't consult ``litellm.api_key`` (the module global Strix sets).
They only check the per-call ``api_key`` kwarg and the
``<PROVIDER>_API_KEY`` env var. The SDK's LitellmModel passes
``api_key=None`` by default, so requests went out with an empty
bearer and DeepSeek (and friends) returned 401.
Mirror the user's LLM_API_KEY into the provider-specific env var
(``DEEPSEEK_API_KEY`` for ``deepseek/...``, ``ANTHROPIC_API_KEY``
for ``anthropic/...``, etc.) using LiteLLM's documented convention.
``os.environ.setdefault`` is used so an explicit user env is never
clobbered. The OpenAI branch was already working via
``set_default_openai_key`` + the existing ``litellm.api_key`` global
fallback.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Brings in 10 commits from main on top of the v1.0.0 branch.
Resolutions:
- Legacy harness files modified on main but deleted in the migration —
kept as deleted: strix/agents/base_agent.py, strix/agents/state.py,
strix/config/config.py, strix/llm/llm.py,
strix/llm/memory_compressor.py, strix/llm/utils.py,
strix/runtime/docker_runtime.py.
- tests/runtime/test_docker_runtime.py — removed; tests dead code.
- strix/skills/vulnerabilities/idor.md and ssrf.md — auto-merged.
- New skills from main kept: header_injection.md, http_request_smuggling.md,
nosql_injection.md, ssti.md.
Five rounds of sweep across the tree. Net ~544 lines removed.
Removed:
- Section-divider banners and one-line section labels (# Display
utilities, # ----- list_requests -----, # CVSS breakdown, etc.).
- Module-level prose docstrings on internal modules. Kept one-line
summaries; trimmed multi-paragraph narration about SDK/Strix
responsibility splits, cache strategies, three-source precedence.
- Internal-helper docstrings that just restate the function name —
caido_api helpers (caido_url, get_client, view_request, etc.),
settings-class one-liners (LLMSettings, RuntimeSettings, ...),
UI helper docstrings.
- Args/Returns blocks on non-LLM-facing internal helpers
(build_strix_agent, render_system_prompt, create_or_reuse,
bootstrap_caido) — kept only the genuinely non-obvious params.
- Internal-history phrasing — "Mirrors main-branch shape",
"pre-SDK harness", "previous lookup matched no attribute".
- Narrative comments inside function bodies that explained what the
next line does, design rationale obvious from the surrounding code,
or "we used to..." asides.
- Trailing periods on every error-string literal across the tool tree.
- Duplicated roundtripTime quirk comment (kept the LLM-facing copy in
tools/proxy/tools.py).
Kept (every one names an upstream bug, vendored-code provenance, or
non-obvious data quirk):
- core/runner.py: SDK replay-with-empty-initial-input + on_agent_end
lifecycle gap.
- runtime/docker_client.py: VERBATIM COPY block of the upstream
_create_container body, pinned to SDK v0.14.6.
- runtime/session_manager.py: NO_PROXY for agent-browser CDP loopback.
- tools/proxy/caido_api.py: generated-pydantic Request.raw quirk,
replay double-history pitfall.
- tools/proxy/tools.py: Caido roundtripTime=0 quirk for proxy
captures.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Four passes of audit-and-patch on the tool surface, condensed.
Tool API shape:
- Todo tools collapse to a single list-based form (one arg per tool,
always a list, no dual-mode validator). Result-field names line up
across the family — created_count / updated_count / marked_count /
deleted_count, and _mark returns a single "marked" key plus the new
status instead of marked_done / marked_pending.
- list_notes splits the overloaded total_count into filtered_count
(matches) and total_count (grand total), matching list_todos. All
three notes mutations now echo total_count and note_id.
- finish_scan drops the machine-code error strings; a single human
"error" key carries the reason on every failure path.
- scope_rules delete echoes a message so the renderer's success
branch has something to surface.
Failure-key unification: every tool now uses {"success": False,
"error": "..."} on failure paths. Touched thinking, web_search,
reporting, and finish. Trailing periods on error strings swept clean
across the whole tool tree.
Tool prompts (docstring re-imports vs main):
- create_vulnerability_report re-imports the CWE reference catalog,
multi-part fix rules, fix_before/fix_after PR-suggestion mechanics,
the COMMON MISTAKES list, the informational-vs-actionable
distinction, and file-path examples.
- web_search re-imports concrete example queries.
- list_sitemap docstring fixed hasDescendants -> has_descendants
(the camelCase reference never matched our snake_case schema).
- create_agent.skills description "Comma-separated" -> "List of".
- factory.py module docstring no longer claims there's no runtime
skill-loading tool. agents_graph module docstring lists stop_agent.
- system_prompt nudges loading the matching skill before guessing
payloads or syntax from memory.
TUI:
- proxy_renderer was reading stale field names from the pre-SDK
schema (requests / total_count / statusCode / matches /
showing_lines); now reads entries / page_info / status_code / hits
/ page+total_lines. Three proxy operations were rendering empty
before this.
- Idle-pane placeholder text trimmed to "Loading...".
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
create_todo / update_todo / mark_todo_done / mark_todo_pending /
delete_todo used to accept either a single-item form (title, todo_id,
…) or a bulk form (todos, updates, todo_ids), reject the call if the
agent set both, and explain the rule in the docstring. The agent kept
tripping the validator. Drop the single-item form everywhere — each
tool now takes one list arg. Single calls just pass a one-item list.
While the API was being reshaped, line the result schemas up:
created_count replaces the lone "count", _mark returns a single
"marked" key plus new_status instead of marked_done / marked_pending,
and list_todos splits the overloaded total_count into filtered_count
(matches) and total_count (grand total) so a filtered call no longer
hides the real size.
Docstrings now spell out each item's fields with required/optional
and the legal status / priority values, plus a worked example.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
exec_command, write_stdin, apply_patch, view_image, load_skill,
list_sitemap, and view_sitemap_entry were falling through to the
generic dict-dumper. They now render in the same visual language as
the rest of the toolset: the terminal pair uses the >_ icon with
pygments bash highlighting; apply_patch and view_image use the file-
edit diamond with colored +/- diff lines and per-language syntax
highlighting; sitemap and load_skill mirror the proxy and skill
patterns already established.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Both backends share session/version/first-run helpers in
strix/telemetry/_common.py and fire from the same four call sites in
strix/interface/main.py and strix/report/state.py. STRIX_TELEMETRY is
the single toggle for both.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Every agent-facing tool now has a corresponding directory: the
strix-implemented ones already do, and the SDK-provided ones
(exec_command/write_stdin shell, apply_patch, view_image) plus the
sandbox-CLI agent-browser get README-only stubs. Each README names the
implementation source, where the tool is wired up, the strix-specific
config it inherits, and the skill that teaches its usage. Listing
strix/tools/ now gives a new reader the full agent toolset at a glance.
The stub dirs intentionally have no __init__.py — they are not Python
packages, just documentation. Nothing in the codebase auto-discovers
strix.tools.* as packages (all imports are explicit), so the stubs
cannot accidentally affect runtime behavior.
The image ships 15 tools (jwt_tool, interactsh-client, arjun, dirsearch,
gospider, wafw00f, retire, eslint, jshint, js-beautify, JS-Snooper,
jsniper.sh, vulnx, ncat, uv) that the always-loaded skills never name
with usage guidance — agents could discover them via the environment
catalog but had no when/how. Add concise mentions in the natural home
for each: jwt_tool in the JWT skill, interactsh-client in the OAST
sections of SSRF/XXE/RCE, arjun in IDOR recon, dirsearch as the broad
alternate in the ffuf skill, gospider + the JS scrapers in katana,
wafw00f next to httpx, retire/eslint/jshint/js-beautify as a new
JavaScript-Side Coverage block in the SAST playbook, uv in python,
vulnx in the deep scan-mode CVE bullet, ncat in a new RCE Tooling
block.
Audit also turned up three real breakages along the way:
- jwt_tool's shebang resolves to /usr/bin/python3 but its dependencies
live in /app/.venv, so every invocation died with
ModuleNotFoundError: ratelimit. Replace the bare symlink with a
wrapper that execs /app/.venv/bin/python against the real script.
- dirsearch's pipx venv ended up with setuptools 82, which dropped
pkg_resources — startup failed before parsing args. Pin the inject
to setuptools<81.
- ESLint's --no-eslintrc flag was removed in v9; the surviving
--no-config-lookup covers it. Drop the dead flag from the SAST
command block.
Also corrected the JS-Snooper / jsniper.sh entry in katana.md — both
take a bare domain and run their own JS discovery internally, not the
JS URLs Katana already harvested.
AGENT_BROWSER_ARGS parser splits on commas, so any flag value
containing one (--disable-features=A,B, --window-size=1920,1080,
--lang=en-US,en) shredded into garbage positionals and Chromium
rejected the launch with "Multiple targets are not supported in
headless mode". Reduce to a comma-separated list of comma-free
flags that keeps the AutomationControlled anti-detection bit.
Default screenshot path now resolves inside the workspace root so
view_image accepts it; entrypoint pre-creates the dir at runtime
(the build-time mkdir is shadowed by the /workspace mount). Skill
examples updated to favor the no-arg form, plus brief fallback
guidance when view_image is unavailable on text-only models and a
viewport-resize note for sites that gate on real desktop dims.
Also drop the stale STRIX_DISABLE_BROWSER doc entry — no code
reference exists.
Sadovoi Grigorii