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>
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.
Main's load_skill tool was deleted during the SDK migration along
with the prompt-mutation pattern it relied on. Re-add the capability
without the mutation: load_skill(skills=[...]) now returns the skill
markdown bodies as a tool result, so the content lands in conversation
history as in-context reference rather than as patched-in system
prompt content. Same source of truth (load_skills + skill files),
same validation (validate_requested_skills) as create_agent.
Tool result format is plain markdown (## Skill: <name> headers joined
with ---), not the <specialized_knowledge> XML wrapping used at
agent-build time. The XML framing was deliberately reserved for
prompt-level privileged context; tool-loaded skills are honestly
labelled as just-fetched reference material.
Close the discovery loop by surfacing the full skill catalog in the
system prompt. Without it the model could only guess skill names —
discovering them via validation errors on misses. Now every agent
sees a categorised <available_skills> block right after the
<specialized_knowledge> block with a short hint pointing at
create_agent / load_skill.
Skills module: factored _iter_user_skill_files() so get_all_skill_names
(set, for validation) and get_available_skills (dict by category,
for the prompt) share one source of truth on what counts as
user-selectable. Internal categories (scan_modes, coordination) stay
excluded from both.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Failure messages were echoing the raw requests-exception text — for an
empty query the model would see "API request failed: 400 Client Error:
Bad Request for url: https://api.perplexity.ai/chat/completions" and
learn the upstream URL, the HTTP status, and the literal word "API"
none of which it has any use for or right to. Same pattern in every
except branch: KeyError leaked internal field names, generic exceptions
leaked library exception text, etc.
Two fixes:
- Pre-flight reject empty/whitespace queries so the trivial misuse case
never hits the network at all and gets a "Query cannot be empty."
result immediately.
- Sanitize every failure path: split RequestException into HTTPError
(4xx → "rejected the query — refine and retry", 5xx → "service
unavailable"), Timeout, ConnectionError, response-shape (KeyError /
IndexError / ValueError), and a generic catch-all. Each path returns
a short actionable message and logs the full traceback via
logger.exception so operator-side observability is preserved. The
model sees no URLs, no status codes, no library exception text.
While in here: the missing-API-key message keeps the env var name
because that's operator-actionable, and the dead "results": [] field
the failure paths used to carry is dropped (success path never had it
either, so the shape was inconsistent).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- view_agent_graph status summary now derives buckets from the canonical
Status literal via get_args, so adding a new status in core.agents
auto-flows into the summary. The previous hardcoded five-bucket list
silently omitted "failed" — buckets stopped summing to total whenever
an agent failed.
- stop_agent rejects targets that are already in a terminal status
(completed / stopped / crashed / failed) with a model-readable error
pointing at view_agent_graph and send_message_to_agent. request_stop
unconditionally overwrites status, so without this guard calling
stop_agent on a completed agent erased the "completed" history.
- StopAgentRenderer added — was falling back to the generic key/value
renderer; the rest of the agents_graph tools have purpose-built ones.
- agent_finish root-rejection payload trimmed from
{success, agent_completed, error, parent_notified} to {success, error}.
The lifecycle gate only reads success+agent_completed and they were
always False/False on this branch, so the extra fields were dead weight.
- wait_for_message renames its top-level outcome field from "status" to
"wait_outcome" — "status" overloaded with the coordinator's agent
status literal (which also has "stopped" as a value, different
meaning). Redundant "agent_waiting" boolean dropped (true iff
wait_outcome == "waiting"). Consumer at factory._wait_tool_parked
updated to match.
- send_message_to_agent now refuses self-send with a pointer at think /
agent_finish / finish_scan instead of looping a message into your
own session.
- SendMessageToAgentRenderer read args.get("agent_id") but the tool's
param is target_agent_id, so the TUI silently never showed the target.
Fixed.
- Restored skill validation lost during the SDK migration: skills
module re-exports get_all_skill_names and validate_requested_skills
(excluding internal scan_modes/coordination categories from the
user-selectable set). create_agent now validates skills before
spawning instead of silently accepting unknown names.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Three gotchas that bite the model once per scan if uncovered:
- HTTPQL has no NOT operator. Naive `NOT req.path.cont:"/static"`
is a parse error. The negated-operator variants (`ne`, `ncont`,
`nlike`, `nregex`) are the only way to negate.
- Strings must be quoted, integers must not. `resp.code.eq:"200"`
parses as a string-vs-int mismatch.
- A bare quoted literal searches both `req.raw` and `resp.raw` —
useful primitive we never surfaced.
All three land in the model-visible tool description.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Re-add list_sitemap and view_sitemap_entry from main, ported to the
new caido-sdk-client layout via raw GraphQL queries (the typed SDK
doesn't expose sitemap operations, but the Caido server still
supports sitemapRootEntries / sitemapDescendantEntries / sitemapEntry).
Wired through caido_api (sandbox-importable helpers), the host-side
@function_tool wrappers, factory _BASE_TOOLS, the system prompt, the
python skill doc, and the public proxy docs.
While threading these through, lock down the output contract across
every proxy tool so the model sees one consistent shape:
- All tools wrap success/failure in {"success": bool, "error"?: str}
- Canonical field names: status_code, length, roundtrip_ms (omitted
when 0), is_tls, has_descendants. snake_case everywhere on output;
camelCase stays only on the input side where it's the GraphQL
schema.
- repeat_request now returns a structured response that matches
list_requests' response_summary shape (parse_raw_response parses
the raw bytes into status_code / length / headers / body), with
body capped at 8KB and a body_truncated flag so the model knows
when to fetch the full body via view_request.
- RepeatRequestRenderer was reading non-existent top-level keys
(status_code, response_time_ms, body) and silently displaying
nothing useful — now reads the structured response shape.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
send_request was a thin wrapper over the Caido Replay API that the model
could replicate with a one-liner `curl` via exec_command. The sandbox's
HTTP_PROXY env captures all such traffic for free, so the tool was
adding bugs (duplicate dispatch, dropped responses) without adding
capability. Removed across factory, tools module, sandbox-importable
caido_api helper, TUI renderer, prompt template, skill doc, and public
docs. repeat_request stays — it operates on captured request IDs with
structured modifications, which curl can't replicate cleanly.
Three caido-sdk-client workarounds that were hitting us through both
send_request and repeat_request:
- replay_send_raw used to pass CreateReplaySessionFromRaw to
sessions.create(), which seeds a stored entry server-side, then
called send() — producing two history rows per call. Empty-create +
send produces one dispatched request.
- The same helper read result.entry.response_raw, an attribute that
doesn't exist on ReplayEntry, so response bytes were silently
dropped. Fixed to walk result.entry.response.raw with proper None
guards.
- get_request_with_client passed include_request_raw / include_response_raw
based on the requested part, but the SDK's generated pydantic models
declare raw as required even though the GraphQL fragment makes it
conditional via @include. Passing False crashed view_request with a
pydantic validation error. Always request both raw bodies; the caller
picks which to surface.
Also wrapped replay.send() in asyncio.wait_for(30s) so a stalled Caido
dispatch (notably loopback targets that don't route cleanly through the
sandbox proxy) fails fast with a model-readable error instead of
hanging the agent until the function_tool 120s budget expires.
Finally, list_requests now omits the roundtrip_ms field when Caido
reports 0 — proxy-captured unscoped traffic consistently reports 0
while scoped/replay traffic carries real measurements, so the absence
of the field is now informative ("Caido didn't measure this") rather
than misleading ("this request took 0ms").
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Notes generated 5-char IDs via a 20-try collision loop while todos
generated 6-char IDs in one shot. Mixed widths across the agent's
view made the two tools look unrelated. Match todo's shape — same
length, same one-shot generation. Collision retry is unnecessary at
scan-scale (a few hundred items vs 16^6 keys).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Reject ambiguous calls in todo tools that previously combined the
single-target params and the bulk-array param (e.g. create_todo with
both `title` and `todos` would silently create N+1 items). Each tool
now errors with a mode-specific hint pointing the model at the
appropriate form. Also drop the meaningless char-count from `think`'s
success message — the model already knows what it wrote.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Two follow-ups from the post-fix audit:
**#1 critical**: ``orchestration/scan.py`` injects the user's new
``--instruction`` into the root's bus inbox via ``bus.send`` on resume,
but ``send`` is one of the deliberately-not-snapshotted high-frequency
mutations. A SIGKILL between that send and the model's first turn
would silently drop the user's new directive. Force a snapshot
immediately after the inject — that's the one specific message we
can't afford to lose, while leaving general ``send`` traffic
unsnapshotted as designed.
**Notes persistence**: ``strix/tools/notes/tools.py`` now mirrors the
todo pattern. ``_notes_storage`` writes through to
``{run_dir}/notes.json`` after every create/update/delete via the
same atomic-tempfile + ``Path.replace`` flow. New
``hydrate_notes_from_disk(run_dir)`` is wired in ``run_strix_scan``
alongside ``hydrate_todos_from_disk`` so a resumed scan recovers the
exact note set the prior process saw, including ``wiki``-category
notes.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Three critical correctness fixes + six TUI/audit/UX fixes from the
parallel-agent audit. All changes verified by an end-to-end smoke
that builds, persists, and re-hydrates state across two simulated
process boundaries.
Critical (resume integrity):
1. ``bus.cancel_descendants_graceful`` now calls ``_maybe_snapshot``
after mutating the ``stopping`` set. Previously, a process crash
between user-initiated graceful-stop and the next finalize lost
the stop signal — respawned agents would run forever instead of
exiting. ``_respawn_subagents`` also gains a guard that skips
agents in ``stopping`` so a previously-cancelled agent is not
resurrected on resume.
2. ``Tracer.hydrate_from_run_dir`` now **raises** on corrupt
``vulnerabilities.json`` instead of swallowing the exception. The
prior behaviour silently reset ``vulnerability_reports`` to empty,
so the next ``add_vulnerability_report`` would allocate ``vuln-0001``
and overwrite the prior MD on disk — silent data loss.
3. ``--instruction`` passed on resume now reaches the model. The CLI
captures whether the user explicitly passed an instruction
(``args.user_explicit_instruction``) before ``_load_resume_state``
loads the persisted one. ``run_strix_scan`` reads
``scan_config["resume_instruction"]`` and, on resume, sends the
new instruction to root's bus inbox before calling
``run_with_continuation`` (which uses ``initial_input=[]`` for SDK
replay). The inject filter surfaces it on the next turn.
4. ``--resume X`` errors loudly when ``scan_state.json`` exists but
``bus.json`` doesn't. Previously this silently fresh-started in
the same dir, confusing the user who explicitly asked to resume.
TUI / audit / UX:
5. ``Tracer.hydrate_from_run_dir`` now reads ``bus.json`` too and
pre-populates ``tracer.agents`` from the snapshot's ``statuses`` /
``names`` / ``parent_of``. Before this, the TUI tree on resume
showed only currently-running agents; completed/crashed children
from the prior run were invisible.
6. ``Tracer.hydrate_from_run_dir`` also seeds ``self._llm_stats`` from
``bus.stats_live + bus.stats_completed`` so the resume's footer
shows cumulative tokens / requests across the prior run plus the
resume segment, instead of resetting to zero.
7. ``Tracer.save_run_data`` now also writes ``run_metadata.json``
(start_time, run_id, run_name, targets, status), and
``hydrate_from_run_dir`` restores ``start_time`` from it. Prior
behaviour reset start_time to ``now()`` on every Tracer init,
breaking the final report's duration calc on resumed scans.
8. Per-agent todos persist to ``{run_dir}/todos.json`` (atomic write
on every CRUD). ``hydrate_todos_from_disk`` (called from
``run_strix_scan``) reloads them so respawned subagents find
their lists intact. Previously, the module-level
``_todos_storage`` was lost on every process restart.
9. ``_load_resume_state`` validates each ``cloned_repo_path`` from
the persisted ``scan_state.json`` still exists on disk. Previously
a deleted clone dir would let the resume proceed with an empty
source tree, with agents silently scanning nothing.
Bonus: ``bus.finalize`` no longer pops ``parent_of`` and ``names``
for finalized agents. Routing protection (don't accept ``send`` to
finalized agents) comes from the ``statuses[id]`` terminal-state
check in ``send`` itself, so dropping those keys was overzealous and
made completed children invisible in ``view_agent_graph`` and the
TUI tree.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
A scan that crashes or is stopped can now be resumed by re-invoking
``strix`` with the same ``--run-name``. Resume is implicit — presence
of ``{run_dir}/bus.json`` triggers it. To force a fresh start, delete
the run dir.
What survives a process restart with the same scan_id:
* Root agent's LLM history — already worked (root SDK SQLiteSession).
* Every non-terminal subagent's LLM history — new. ``create_agent``
now opens SQLiteSession(session_id=child_id,
db_path={run_dir}/sessions/{child_id}.db) per child and passes it
to ``run_with_continuation``.
* Bus topology — new. ``AgentMessageBus`` gains snapshot/restore/
_maybe_snapshot async methods plus a ``metadata`` field that holds
per-agent {task, skills, is_whitebox, scan_mode, diff_scope}.
``register``, ``finalize``, ``park``, and ``mark_llm_failed`` each
call ``_maybe_snapshot`` to atomically persist the bus to
{run_dir}/bus.json (tempfile + Path.replace).
* Vulnerability reports — new. ``ScanArtifactWriter._write_
vulnerabilities`` now also writes ``vulnerabilities.json``
(atomic). ``Tracer.hydrate_from_run_dir`` reads it on resume so
new vuln-NNNN ids don't collide with prior on-disk files.
What does not survive: the sandbox container itself (fresh per
process), so ``/workspace/scratch`` and Caido state are lost.
``/workspace/sources`` re-mounts from the host so source code is
unchanged.
``orchestration/scan.py:run_strix_scan`` does the actual resume:
1. Resolve run_dir up front; if bus.json exists it's a resume.
2. Acquire {run_dir}/.lock (fcntl.flock) so a second strix process
can't run concurrently on the same scan_id.
3. ``bus.set_snapshot_path(...)``, ``tracer.hydrate_from_run_dir()``.
4. On resume: load + bus.restore, find root_id from snapshot (the
agent with parent_of[id] is None), spawn the sandbox, skip the
root's bus.register (already in snapshot).
5. ``_respawn_subagents`` walks every agent with status in
running/waiting/llm_failed: reopens its SQLiteSession, rebuilds
the child agent via the captured factory, builds run config /
context, asyncio.create_task the run with initial_input=[] so
the SDK replays from session. Per-child failure (missing/corrupt
DB, factory raises) finalizes that child as crashed and continues.
6. Open root SQLiteSession at the same path, run the root with
initial_input=[] on resume (or the formatted root task on a
fresh run), and let SDK replay drive the next turn.
7. ``finally``: close every per-agent session, take a final
snapshot, tear down sandbox, release the lock.
HARNESS_WIKI.md updated with the new run-dir layout (sessions/,
bus.json, vulnerabilities.json, .lock) and the resume contract.
Net: +500 LoC across 7 files. No new deps.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The notes tool no longer touches disk. ``_notes_storage`` lives in
memory for the lifetime of one scan process, shared across every
agent in that process via the existing RLock. Process exit clears
the lot — no notes.jsonl event log, no wiki/<slug>.md Markdown
rendering, no replay-on-startup hydration.
Removed ~10 internal helpers (``_get_run_dir``,
``_get_notes_jsonl_path``, ``_append_note_event``,
``_load_notes_from_jsonl``, ``_ensure_notes_loaded``,
``_persist_wiki_note``, ``_remove_wiki_note``,
``_get_wiki_directory``, ``_get_wiki_note_path``,
``_sanitize_wiki_title``) plus the ``_loaded_notes_run_dir`` module
state, ``wiki_filename`` per-note field, and the ``OSError`` branches
that only existed for the wiki write path.
The ``wiki`` category is preserved as a free-form long-form bucket;
it just no longer has any special persistence behaviour.
Skill prompts scrubbed of every "shared wiki memory" / "repo wiki" /
"append a delta before agent_finish" instruction:
``coordination/source_aware_whitebox.md``,
``custom/source_aware_sast.md``,
``scan_modes/{quick,standard,deep}.md``, plus the WHITE-BOX TESTING
block in ``agents/prompts/system_prompt.jinja``.
HARNESS_WIKI.md updated to drop the wiki-as-shared-knowledge-base
description, the per-run output-tree references to ``notes/notes.jsonl``
and ``wiki/{note_id}-{slug}.md``, and the ``is_whitebox`` toggle prose.
Net: -178 LoC in notes/tools.py, -45 LoC across skills/system_prompt
and the wiki doc. The notes tool surface (5 ``@function_tool``s) is
unchanged for the agent.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Five gaps from the post-implementation audit, closed:
1. **SDK logger captured.** The openai-agents SDK uses
``logging.getLogger("openai.agents")`` for its own lifecycle events
(Runner.run starts, tool dispatch, model retries, exceptions).
Previous setup only attached handlers to the ``strix`` root, so
SDK-internal events were dropped. Tracked-roots tuple now covers
both, with the same FileHandler/StreamHandler/Filter chain.
2. **Proxy tool exception tracebacks.** Every ``@function_tool`` in
``strix/tools/proxy/tools.py`` returns a JSON error to the LLM via
the ``_err(name, exc)`` helper. The tracebacks were silently
formatted away — the LLM saw the message, the human reading the
log saw nothing. ``_err`` now emits ``logger.exception(...)``
covering all five tools at once.
3. **CLI bootstrap.** ``strix/interface/main.py`` had its module
``logger`` removed by the previous commit and was emitting nothing.
Restored, plus log lines for env validation, docker check, LLM
warm-up, and image pull (debug for already-present, info for
pull, exception for failures).
4. **Docker client.** ``strix/runtime/docker_client.py`` had no
logger. Container creation now logs caps + exposed ports at DEBUG
and the resulting container id at INFO.
5. **PostHog telemetry.** ``strix/telemetry/posthog.py`` had no
logger. Now logs send success/failure at DEBUG, version-detection
failures at DEBUG, and disabled-skip at DEBUG (so the log shows
when telemetry is off, instead of being silent about it).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Every scan now writes a complete log file at ``{run_dir}/strix.log``
captured from the moment ``run_dir`` is resolved through teardown.
Stdlib ``logging`` only — no parallel framework.
New ``strix/telemetry/logging.py``:
* ``setup_scan_logging(run_dir, debug=)`` attaches a ``FileHandler``
(DEBUG, all ``strix.*``) plus a ``StreamHandler`` (ERROR by
default; DEBUG via ``STRIX_DEBUG=1``).
* ``ContextVar``-backed ``scan_id`` and ``agent_id`` injected by a
``Filter`` so every line is auto-tagged across asyncio tasks
without callers passing them explicitly.
* Third-party noise (``httpx``, ``litellm``, ``openai``,
``anthropic``, ``urllib3``, ``httpcore``) capped at WARNING.
* Returns a teardown handle for ``finally`` cleanup.
Wiring:
* ``orchestration/scan.py`` calls ``setup_scan_logging`` once per
scan after ``run_dir`` resolves; sets scan_id; tears down in
``finally``. Adds INFO logs for sandbox bring-up + scan
start/end.
* ``orchestration/hooks.py`` sets/clears ``agent_id`` ContextVar in
``on_agent_start`` / ``on_agent_end`` and emits INFO for agent
lifecycle, DEBUG for every tool start/end and LLM call.
* ``interface/main.py`` drops the ``setLevel(ERROR)`` silencer.
Coverage expanded across ~20 files (orchestration, agents, runtime,
llm, tools, interface, config, skills) with INFO for lifecycle and
DEBUG for verbose detail. Per the system instructions in
``logger.warning(f"…{e}")`` were converted to module logger calls.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Restores the legacy persistent-IPython tool's *ergonomics* (proxy
helpers pre-bound, structured stdout/stderr/error returns) without the
in-container daemon: each call ships ``strix.tools.proxy._calls`` source
into ``/tmp`` alongside a per-call driver, runs ``python3 -u`` against
it, and parses a sentinel-delimited JSON payload back from stdout. The
driver fetches its own guest token from Caido at ``localhost:48080``
and binds ``list_requests`` / ``view_request`` / ``send_request`` /
``repeat_request`` / ``scope_rules`` to that client; user code runs
inside an ``async def`` wrapper so top-level ``await`` works.
The proxy SDK call sequences live in one file —
``strix/tools/proxy/_calls.py`` — and are reused by both the host-side
``@function_tool`` wrappers (which add JSON serialization for the LLM)
and the in-container kernel (which exposes the bare async functions).
No code duplication; the helper logic itself is host-shipped, so
tweaking the proxy helpers does not require an image rebuild.
Image: a single ``pip install caido-sdk-client`` line so the driver's
``import caido_sdk_client`` resolves. Skill ``tooling/python`` is
always-loaded alongside ``tooling/agent_browser``.
Trade-off accepted: state does not persist across calls (no kernel).
For multi-step workflows the agent combines into one ``code`` block or
writes a script to ``/workspace/scratch/`` and runs via
``exec_command``. If a workflow surfaces that genuinely needs
persistence, the same tool surface migrates to a kernel-backed
executor without changing the LLM contract.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Three top-level files that didn't earn their place:
- ``strix/io/scan_artifacts.py`` had a single consumer (the Tracer);
collapsing it into ``strix/telemetry/`` puts it next to that consumer.
``strix/io/`` is gone.
- ``strix/run_config_factory.py`` held two helpers that didn't earn the
factoring. ``make_agent_context`` was a 17-line dict-spelling function
whose argument names were identical to its dict keys — replaced with
inline dict literals at the two call sites. ``make_run_config`` had
enough RunConfig assembly logic to justify a helper, but with only
two callers (root scan + ``create_agent``) inlining is cleaner than
keeping a top-level file. ``DEFAULT_RETRY`` moves to
``strix/llm/retry.py`` next to its other LLM-policy peers; the dead
``STRIX_DEFAULT_MAX_TURNS`` constant is dropped.
- ``strix/entry.py`` is a misnomer — it isn't *the* entry point (that's
``strix/interface/main.py`` for the CLI), it's the per-scan bring-up
driver: build the bus, bring up the sandbox, build the root agent +
child factory, format the scope-context block, register root in bus,
open SQLiteSession, hand off to ``run_with_continuation``. That all
lives next to its peers in ``strix/orchestration/`` now, renamed to
``scan.py`` so the role is obvious.
No behavior change. Net -125 LoC.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Audit flagged that legacy ``finish_scan`` had a code-level guard
(``_check_active_agents``) that refused completion if any subagent was
still running or stopping. Restoring it as code would be defensive
mid-stream cancellation we don't actually want — the agent should
choose whether to wait, message, or stop each child.
Lift the responsibility to the prompt instead: docstring now opens
with a numbered pre-flight checklist that requires the agent to
``view_agent_graph`` first and refuses self-permission to call
``finish_scan`` while any peer is in ``running`` / ``waiting`` /
``llm_failed``. The model sees this as part of the tool's schema and
treats it as a hard rule (matches our pattern for similar
constraints).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Final pass after re-audit. Three sub-specs landed:
**XML simplification** — the legacy XML envelopes were prompt-engineering
ceremony, not parser primitives (the SDK uses native tool-calling). Drop
the verbose wrappers in favor of one-liner labeled headers. Side benefit:
fixes the unescaped-content XML-injection bug the audit caught (peer
content containing ``</content>`` no longer breaks the wrapper).
- ``_format_inter_agent_message``: ``<inter_agent_message><sender>...
<content>...`` 9-line XML → ``[Message from {name} ({id}) | type=... |
priority=...]\n{content}``.
- ``_render_completion_report``: ``<agent_completion_report><agent_info>
...<results>...`` XML → human-readable structured text with section
headers and bulleted lists.
- ``inherited_context``: ``<inherited_context_from_parent>...`` →
``== Inherited context from parent (background only) ==``.
**MG1: TUI stop-agent uses graceful cancel.** ``tui.py`` was calling
``bus.cancel_descendants`` (hard, ``task.cancel()`` mid-stream) for the
stop-agent button. Switched to ``bus.cancel_descendants_graceful``, which
uses ``RunResultStreaming.cancel(mode="after_turn")`` to let each agent
finish its current turn (and save to session) before honoring the cancel.
The hard path remains in ``entry.py`` for KeyboardInterrupt where
graceful isn't possible.
**MG2: Document hook lock-free stats mutation.** Added a comment in
``hooks.on_llm_start`` explaining why ``warned_85`` / ``warned_final``
are mutated lock-free: SDK serializes ``on_llm_start`` per agent, so this
hook is the sole writer to those keys; ``record_usage`` only writes
disjoint keys (in/out/cached/calls).
**AG3: Auto-load ``coordination/root_agent`` skill for the root.**
Legacy auto-loaded the orchestration-guidance skill for root agents
only. Threaded ``is_root`` through ``render_system_prompt`` →
``_resolve_skills``; root agents now get the skill, children don't.
Skipped (per user direction): whitebox-wiki integration (CG2-4) — the
auto-injection / auto-update of the shared repo wiki was a pre-migration
feature; user opted not to restore it.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Audit found 8 behavioral gaps between post-migration and the legacy
``BaseAgent.agent_loop``. All 8 are now closed using SDK-native
primitives — no custom workarounds, no shadow state machines.
What was broken / different:
- G1: ``inherit_context`` was dead code; children always started fresh.
- G2: TUI user message couldn't interrupt an in-flight LLM/tool turn.
- G3: ``llm_failed`` state never set; hard failures propagated as crashes.
- G4: No graceful ``stop_agent`` tool.
- G5: Parked subagents waited forever (no auto-resume timeout).
- G6: Inter-agent messages used a plain header instead of legacy XML.
- G7: Completion reports used JSON instead of legacy XML.
- G11/G12: Turn counter reset per cycle; budget warnings could re-fire.
What we did:
Bus extensions (``orchestration/bus.py``):
- ``streams`` registry + ``attach_stream`` ctx manager + ``request_interrupt``
for SDK-native ``RunResultStreaming.cancel(mode="after_turn")``.
- ``mark_llm_failed`` + ``wait_for_user_message`` (filtered: only ``from="user"``
satisfies; peer messages don't unstick a stuck model).
- ``stopping: set[str]`` for graceful programmatic exit.
- ``cancel_descendants_graceful`` — leaves-first via ``request_interrupt``.
- ``record_usage`` increments ``calls`` unconditionally so it doubles as the
per-agent-lifetime turn counter (legacy ``state.iteration`` parity).
- ``warned_85`` / ``warned_final`` flags on ``stats_live`` for once-fire
budget warnings.
Run loop rewrite (``orchestration/run_loop.py``):
- ``Runner.run`` → ``Runner.run_streamed`` with ``bus.attach_stream`` so
cancel has a target. Catch ``(AgentsException, APIError)`` after retries
exhaust; in interactive mode call ``mark_llm_failed`` + wait for user.
- ``UserError`` / ``MaxTurnsExceeded`` / ``CancelledError`` propagate.
- Outer loop: ``asyncio.wait_for(bus.wait_for_message, timeout=300)`` for
interactive subagents (root waits forever). ``TimeoutError`` injects
``"Waiting timeout reached. Resuming execution."``.
- Honors ``bus.stopping`` at top of each iteration.
Hooks (``orchestration/hooks.py``):
- Counter source moved from per-cycle ``ctx["turn_count"]`` to
per-lifetime ``bus.stats_live[agent_id]["calls"]``.
- Warnings guarded by once-flags — exactly-once across all cycles.
Filter (``orchestration/filter.py``):
- Restored legacy ``<inter_agent_message>`` XML envelope with the
``<delivery_notice>DO NOT echo back</delivery_notice>`` instruction.
Agents-graph (``tools/agents_graph/tools.py``):
- G1: ``create_agent`` reads ``ctx.turn_input`` (SDK populates it before
tool execution at ``run_internal/turn_resolution.py:806``). Wraps as
one ``<inherited_context_from_parent>`` block.
- G7: ``agent_finish`` emits the legacy ``<agent_completion_report>``
XML. ``child_ctx["task"] = task`` threaded so the report echoes the
original task.
- G4: New ``stop_agent`` tool — refuses self-stop, refuses already-
finalized targets, ``cascade=True`` uses ``cancel_descendants_graceful``.
TUI (``interface/tui.py``):
- ``_send_user_message`` schedules ``bus.send`` AND
``bus.request_interrupt(target, mode="after_turn")`` — SDK finishes
current turn cleanly, next cycle picks up the user's message.
Factory (``agents/factory.py``):
- Registered ``stop_agent`` in ``_BASE_TOOLS``.
Out of scope:
- G8 (``[ABORTED BY USER]`` marker) is auto-resolved by G2 — the SDK
saves the full assistant message before honoring
``cancel(mode="after_turn")``, so partial content is preserved in the
session.
Verified all bus behaviors with a smoke test. Lint at baseline.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Top-level ``strix/run_loop.py`` was an orphan — it owns the multi-agent
continuation loop, which is exactly the orchestration layer's job.
Moves it into ``strix/orchestration/run_loop.py`` next to the bus,
hooks, and filter — they all glue ``Runner.run`` to bus state.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The previous commit only kept the root agent alive across cycles. But
``interactive`` propagates to children via ``make_child_factory``, and
the legacy harness's continuation loop applied to every interactive
agent in the tree — children also stayed alive after ``agent_finish``,
ready to receive follow-up messages from the parent or siblings.
Lift the demo-loop pattern out of ``entry.run_strix_scan`` into a
shared helper :func:`strix.run_loop.run_with_continuation` and use it
at both call sites:
- ``entry.run_strix_scan`` for the root agent.
- ``tools.agents_graph.tools.create_agent`` for child agents — the
``asyncio.create_task(Runner.run(...))`` becomes
``asyncio.create_task(run_with_continuation(...))``.
``StrixOrchestrationHooks.on_agent_end`` drops the ``parent_id is None``
constraint — any interactive agent parks instead of finalizing.
Children that crash still finalize so parents stop waiting on them.
Cancellation propagates correctly: ``bus.cancel_descendants`` cancels
the task; ``run_with_continuation``'s ``await bus.wait_for_message``
catches ``CancelledError`` and returns the last result.
Lint at baseline.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Pre-migration ``BaseAgent.agent_loop`` ran forever in interactive mode,
re-entering a "waiting state" after each finish-tool call so user
follow-ups could keep the conversation going. Post-migration our
``Runner.run`` returned on ``StopAtTools(finish_scan)`` and the user's
next chat message had no listener — silent dead-end.
Restore the legacy "agent never dies" semantics using the SDK's
canonical demo-loop pattern (``agents/repl.py:run_demo_loop``):
- Add ``AgentMessageBus.wait_for_message(agent_id)`` — blocks until
an inbox is non-empty. Backed by a per-agent ``asyncio.Event``
fired from ``send``.
- Add ``AgentMessageBus.park(agent_id)`` — sets status to ``waiting``
without finalizing (inbox + tree edges + name preserved). Lets
``send`` keep accepting messages between cycles.
- Plumb ``interactive`` through ``make_agent_context`` and the
``create_agent`` graph tool (children inherit).
- ``StrixOrchestrationHooks.on_agent_end`` parks the root agent
instead of finalizing when ``interactive=True`` and the run
completed cleanly. Resets ``agent_finish_called`` /
``turn_count`` for the next cycle.
- ``entry.run_strix_scan`` adds an outer loop in interactive mode:
after ``Runner.run`` returns, ``await bus.wait_for_message(root_id)``,
drain pending user messages, and re-invoke ``Runner.run``. SQLite
session preserves prior conversation across cycles.
For non-interactive (CLI) mode: unchanged — single ``Runner.run``,
return.
Verified bus behaviors: wait returns immediately on pre-existing
message, blocks then wakes on send, ``park`` keeps agent send-able,
``finalize`` evicts. Lint at baseline (3 ruff / 69 mypy).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
``agent_finish`` was setting ``inner[\"agent_finish_called\"] = True``
at the top of its body, but ``StrixOrchestrationHooks.on_tool_end``
already does this for ``agent_finish`` and ``finish_scan`` after the
tool returns. Doing it twice was harmless but suggested the flag's
ownership was ambiguous; the hook is the single source of truth.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
``"anthropic/claude-sonnet-4-6"`` was duplicated as a kwarg default in
5 places (``run_strix_scan``, ``make_run_config``, ``make_agent_context``,
and twice in ``agents_graph.create_agent``'s ``inner.get(..., default)``
calls). The default was actually dead code: ``validate_environment``
requires ``STRIX_LLM`` to be set before any scan starts, and the CLI/TUI
callers don't pass ``model=`` themselves.
Replaced with a single resolution in ``run_strix_scan``:
resolved_model = model or load_settings().llm.model
if not resolved_model:
raise RuntimeError("No LLM model configured. ...")
then propagated explicitly to ``make_agent_context`` and
``make_run_config``. Both lose their string defaults — ``model`` is now
a required kwarg. The graph tool's ``inner.get("model", "...")`` is
``inner["model"]``: the parent context guarantees it's set.
Drive-by: ``run_strix_scan`` docstring still listed ``is_whitebox`` as
a ``scan_config`` key — stale since ``1e641e5`` derived it from
``targets`` instead. Updated.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Replaces 200+ lines of bespoke env-loader / persist / change-detection
machinery with ``pydantic_settings.BaseSettings`` (already a transitive
of ``openai-agents → mcp``, no new direct dep).
What was wrong with ``Config``:
- 14 knobs flat in one namespace, weak grouping by comment-block.
- ``Config._applied_from_default`` and ``Config._config_file_override``
were externally mutated from ``interface/main.py:532-534``. Private
members were part of the public contract.
- Stringly-typed values: every caller had to coerce
(``int(Config.get("llm_timeout") or "300")``,
``... not in {"0", "false", "no", "off"}``).
- Dead knob: ``strix_llm_max_retries`` declared, persisted, listed in
``_LLM_CANONICAL_NAMES`` — zero readers (``DEFAULT_RETRY``
hardcodes ``max_retries=5``). Dropped.
- ``_LLM_CANONICAL_NAMES`` tuple maintained alongside class vars —
duplicate source of truth.
- ``_tracked_names()`` introspected ``vars(cls).items()`` filtered on
``(v is None or isinstance(v, str))`` — fragile.
- Awkward path: ``strix/config/config.py`` inside ``strix/config/``
with ``__init__.py`` just re-exporting.
- Dual access for the same fact: ``web_search`` read
``os.getenv("PERPLEXITY_API_KEY")`` while ``main.py`` read
``Config.get("perplexity_api_key")``.
New shape:
- ``strix/config/settings.py`` — typed dataclass tree:
``Settings.{llm,runtime,telemetry,integrations}``. Each sub-model is
its own ``BaseSettings`` so it reads env independently. Field-level
``alias=`` and ``validation_alias=AliasChoices(...)`` mirror the
existing flat env-var names — user-facing env contract is unchanged.
Bool fields auto-parse ``"0"``/``"false"``/``"no"``/``"off"``;
int fields auto-coerce.
- ``strix/config/loader.py`` — thin ``load_settings()``,
``apply_config_override(path)``, ``persist_current()`` with module
cache. JSON file reader walks aliases to populate sub-models, dropping
entries already covered by env (so env still wins).
- 13 callsites migrated from ``Config.get("...")`` to
``load_settings().<group>.<field>``.
- ``posthog._is_enabled()`` collapses to one line.
- ``--config <path>`` flow simplified: one
``apply_config_override(...)`` call replaces three lines of
class-private mutation.
Drive-by — drop ``is_whitebox`` from ``scan_config`` dict:
- It was being derived as ``bool(args.local_sources)`` in three places
(``cli.py``, ``tui.py``, ``main.py``) and stuffed into the dict for
``entry.py`` to read back. The fact is fully derivable from
``scan_config["targets"]`` — any target with ``type == "local_code"``.
- New helper ``is_whitebox_scan(targets)`` in ``interface/utils.py``
alongside the other target-classification utilities.
- ``entry.py`` computes once; ``main.py``'s posthog start uses the same
helper. Triplicate derivation gone.
Verified: ruff at baseline (3), mypy at baseline (69). Six smoke tests
pass — defaults / JSON-only / env-wins-over-JSON / alias-chain
fallback / bool parsing / ``is_whitebox_scan``.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
``check_duplicate`` was calling ``litellm.completion(...)`` directly
via ``resolve_llm_config()``, bypassing every layer the main agent
loop runs through:
- :class:`MultiProvider` (so ``anthropic/...`` aliases never went
through :class:`AnthropicCachingLitellmModel` and missed the
``cache_control`` patching on the system prompt — 4x cost on
repeated dedupe calls within the same scan).
- :data:`DEFAULT_RETRY` (no retry on 429s / network blips — the
caller's broad except-and-fallback was hiding this).
Switch to the SDK's :meth:`Model.get_response` directly: same model
selection, same retry policy, same cache wrapper. Extract assistant
text from ``ModelResponse.output`` via the canonical
``ResponseOutputMessage`` walk.
``check_duplicate`` is now async — drops the ``asyncio.to_thread``
indirection in ``_do_create``. Validation logic is fast-sync; running
it on the event loop is fine.
Drive-by: rename ``_DEFAULT_RETRY`` → ``DEFAULT_RETRY`` in
``run_config_factory`` so the dedupe path can reuse the same constant
without reaching into a private name.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
``@strix_tool`` was passing through every kwarg to ``@function_tool``
with the same defaults — zero Strix-specific value-add. The docstring
also still claimed terminal/browser/python tools opted into
``timeout_behavior="raise_exception"``, but those tools were all
deleted in the recent migrations.
- Replace 30 ``@strix_tool(...)`` callsites with ``@function_tool(...)``.
- Inline ``dump_tool_result(x)`` as ``json.dumps(x, ensure_ascii=False,
default=str)`` at all 64 callsites — no helper.
- Delete ``strix/tools/_decorator.py``.
Drive-by: gut dead package re-exports.
- ``strix/{agents,orchestration,tools}/__init__.py`` re-exported
symbols nobody imports via the package — every consumer uses deep
paths (``from strix.agents.factory import build_strix_agent``).
- The 8 ``strix/tools/<sub>/__init__.py`` re-exports only fed the
splat ``from .agents_graph import *`` etc. in the parent package
init, which is also gone now.
- Reduced to docstrings (or empty) so ``import strix.tools`` doesn't
drag every tool's transitive deps in eagerly.
Drive-by: drop dead helpers in ``runtime.session_manager``
(``cached_scan_ids``, ``_reset_cache_for_tests``) — zero callers since
``tests/`` was nuked in ``a6d578c``.
Verified all tool timeouts preserved (think=10, list_requests=120,
finish_scan=60, web_search=330) and ruff/mypy at baseline.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The split between ``strix/sandbox/`` and ``strix/runtime/`` was
artificial — both were managing the same backend. ``strix/sandbox/``
also collided uncomfortably with the SDK's ``agents.sandbox.*``
namespace. ``runtime/`` (which matches ``STRIX_RUNTIME_BACKEND``) is
the canonical home for everything Docker / Daytona / K8s lifecycle.
While merging, also rip out two pieces of Docker-specific coupling:
- ``caido_bootstrap`` was POSTing ``loginAsGuest`` from the host via
``aiohttp`` to ``http://127.0.0.1:{forwarded_port}``. That assumed
Docker port forwarding; Daytona / K8s expose ports differently.
Now we ``session.exec`` curl from *inside* the container — the
SDK's runtime-agnostic exec primitive — so any backend works as
long as it implements ``exec``. The host-side Caido ``Client``
still uses the runtime's exposed-port URL for post-bootstrap calls,
but that goes through the SDK's own ``resolve_exposed_port``
abstraction (also runtime-agnostic).
- The bootstrap retry loop now doubles as the readiness probe, so
``healthcheck.wait_for_tcp_ready`` (and the entire
``healthcheck.py`` module) goes away.
Drive-by simplification: drop ``caido_host_port`` plumbing entirely.
It was only piped through ``make_agent_context`` → child contexts
without ever being read; only ``caido_client`` is consumed.
Drops ``aiohttp`` runtime dep (it stays only as a transitive of the
Caido SDK).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Combined commits 2+3 of the migration plan because the FastAPI sidecar
removal in commit 2 broke ``browser_action`` (which lived in the
sidecar); they have to land together.
Sandbox tool layer (commit 2 piece):
- ``build_strix_agent`` now returns a ``SandboxAgent`` with
``capabilities=[Filesystem(), Shell()]``. The SDK runtime binds the
capabilities to the live sandbox session per-run; agents get
``exec_command``, ``write_stdin``, ``apply_patch``, ``view_image``
function tools auto-merged into their tool list. Plain ``Agent``
short-circuits capability binding (``agents/sandbox/runtime.py:190``).
- Drop ``Compaction`` from the default capability set — it's
OpenAI-Responses-API-only and useless for our litellm-routed
Anthropic setup.
- Delete the entire custom in-container tool layer:
- ``strix/tools/terminal/`` (5 files, 748 LoC libtmux)
- ``strix/tools/file_edit/`` (3 files, 276 LoC)
- ``strix/tools/python/`` (5 files, 459 LoC)
- ``strix/runtime/tool_server.py`` (163 LoC FastAPI sidecar)
- ``strix/tools/_sandbox_dispatch.py`` (117 LoC)
- ``strix/tools/registry.py`` (109 LoC)
- ``strix/tools/context.py`` (12 LoC)
- Drop the corresponding TUI renderers (``terminal_renderer.py``,
``file_edit_renderer.py``, ``python_renderer.py``) and update
``interface/tool_components/__init__.py``.
Browser → agent-browser CLI (commit 3 piece):
- Install ``agent-browser@0.26.0`` globally in the Dockerfile right
after the existing ``npm install -g`` block. Run
``agent-browser install --with-deps`` (apt, root) and
``agent-browser install`` (Chrome download, pentester) +
``agent-browser doctor --offline --quick`` smoke test.
- Drop the explicit Playwright system-deps apt list (replaced by
``--with-deps``) and ``RUN .venv/bin/python -m playwright install
chromium``.
- Vendor ``agent-browser/skill-data/core/SKILL.md`` →
``strix/skills/tooling/agent_browser.md`` (476 lines). Adapt
frontmatter to Strix format; strip the install/Quickstart and the
``agent-browser skills get electron|slack|...`` specialized-skills
block; add the "Caido proxy is wired via env vars; do not pass
``--proxy``" note.
- ``_resolve_skills`` now eagerly loads ``tooling/agent_browser`` for
every agent (matches the previous unconditional ``browser_action``
in ``_BASE_TOOLS``).
- Delete ``strix/tools/browser/`` (5 files, 1338 LoC) and the
``browser_renderer.py`` TUI render.
Sandbox plumbing:
- Drop ``bearer`` token, ``tool_server_host_port`` resolution + bundle
keys, ``TOOL_SERVER_TOKEN``/``TOOL_SERVER_PORT``/
``STRIX_SANDBOX_EXECUTION_TIMEOUT`` from the manifest env in
``session_manager.create_or_reuse``. Caido proxy env vars
(``http_proxy``, ``https_proxy``, ``ALL_PROXY``) stay; manifest
applies them to every ``docker exec``-spawned process.
- Drop ``sandbox_token`` and ``tool_server_host_port`` params from
``make_agent_context`` and the ``create_agent`` graph tool.
- Drop the tool-server health-check from ``entry.py`` (only Caido's
``wait_for_tcp_ready`` remains).
- ``docker-entrypoint.sh``: delete the ~30 line
``Starting tool server...`` block (sudo + uvicorn launch + curl
/health poll). Add ``NO_PROXY=localhost,127.0.0.1`` to
``/etc/profile.d/proxy.sh`` and ``/etc/environment`` so the
agent-browser daemon's CDP traffic on localhost isn't routed
through Caido.
pyproject.toml:
- ``[project.optional-dependencies] sandbox = []`` (every member of
the previous list — fastapi, uvicorn, ipython, openhands-aci,
playwright, libtmux — is gone with the sidecar).
- Drop ``numpydoc.*``, ``IPython.*``, ``openhands_aci.*``,
``playwright.*``, ``uvicorn.*``, ``pyte.*``, ``libtmux.*`` from
the missing-imports module list.
- Drop the per-file ruff ignores for the deleted modules.
Net delta: −5512 LoC. ruff drops to 3 errors (was 21 baseline). mypy
falls to 69 errors over 3 files (was 84 over 8 — the drop comes from
deleting the modules with the worst untyped-import problems).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Drop our 797-LoC manual GraphQL ``ProxyManager`` and the in-container
sandbox dispatch. Caido goes host-side via the official async Python
SDK. The Caido CLI still runs as a sidecar in the container — only the
control-plane moves.
Bootstrap moves host-side:
- New ``strix/sandbox/caido_bootstrap.py``: ``loginAsGuest`` via
aiohttp (5 retries), then ``client.project.create(temporary=True)``
+ ``client.project.select(...)``, then return the connected
``caido_sdk_client.Client``. Drop the equivalent bash from
``docker-entrypoint.sh`` (~60 lines of curl + jq).
- ``entry.py`` calls ``bootstrap_caido_client`` after the
``wait_for_tcp_ready`` healthcheck, stashes the client in the bundle
and threads it through ``make_agent_context(caido_client=...)``.
``agents_graph.create_agent`` propagates the same client to children.
- ``session_manager.cleanup`` ``await``s ``client.aclose()`` before
tearing down the container.
- Drop ``CAIDO_PORT`` from the manifest env (only the in-container
ProxyManager read it) and ``CAIDO_API_TOKEN`` from the entrypoint's
``/etc/profile.d/proxy.sh`` + ``/etc/environment`` heredocs.
Tools (``strix/tools/proxy/tools.py``):
- ``list_requests`` → ``client.request.list().filter().first().after()``
with ascending/descending order. **Pagination changes from
start_page/end_page (1-indexed) to first/after cursors** matching the
SDK's native shape; response includes ``page_info.end_cursor`` for
the model to thread.
- ``view_request`` → ``client.request.get(id, RequestGetOptions(...))``;
decode raw bytes locally; existing regex-search and line-pagination
modes preserved.
- ``send_request`` → synthesize raw HTTP bytes, parse URL into
``ConnectionInfoInput(host, port, is_tls)``, create a replay session
via ``client.replay.sessions.create(CreateReplaySessionFromRaw(...))``,
then ``client.replay.send(session_id, ReplaySendOptions(...))``.
- ``repeat_request`` → ``client.request.get(id, request_raw=True)`` →
port the existing parse/_apply_modifications/build helpers verbatim →
send via the same replay flow as ``send_request``.
- ``scope_rules`` → direct mapping to ``client.scope.{list, get, create,
update, delete}``.
- **Drop ``list_sitemap`` + ``view_sitemap_entry``** — the official SDK
has no sitemap module. The model uses HTTPQL filters
(``req.host.eq:"X" AND req.path.cont:"/api/"``) for the same
drill-down workflow.
Deletions:
- ``strix/tools/proxy/proxy_manager.py`` (797 LoC)
- ``strix/tools/proxy/proxy_actions.py`` (113 LoC)
- The 6-line proxy_actions pre-import in ``python_instance.py``
(broken once proxy_actions is gone; that file is queued for deletion
in commit 2 anyway).
Deps:
- Add ``caido-sdk-client>=0.2.0`` and ``aiohttp>=3.10.0`` to runtime
``[project] dependencies``.
- Drop ``gql[requests]>=3.5.3`` from ``[project.optional-dependencies]
sandbox`` — only the in-container ProxyManager used the sync transport
variant; the SDK pulls in ``gql[aiohttp]`` transitively for us.
- ``[[tool.mypy.overrides]]``: add ``caido_sdk_client.*`` and
``aiohttp.*`` to the missing-imports list with
``disable_error_code=["import-untyped"]`` (neither ships ``py.typed``).
- ``[tool.ruff.lint.per-file-ignores]``: bump the proxy/tools.py
ignore to also include ``PLR0911`` (the scope_rules action dispatcher
has many short-circuit returns).
ruff drops from 21 → 12 errors; mypy moves from 82 → 84 (the +2 are in
already-flaky files unrelated to this change). All touched files mypy
clean.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The custom ``Capability`` subclass was 207 LoC bundling four tiny
concerns (env-var injection, tool exposure, system-prompt block,
healthcheck) — and three of them were dead code: the SDK's
``SandboxRunConfig`` doesn't accept capabilities, so
``process_manifest``, ``tools()``, and ``instructions()`` were never
called. Only ``bind()`` ran, because we invoked it manually.
Replace each piece with the obvious direct equivalent:
- **Env vars**: inject ``http_proxy`` / ``https_proxy`` / ``ALL_PROXY``
directly into the manifest in ``session_manager.create_or_reuse``.
This *also fixes a latent bug* — the proxy env vars in
``CaidoCapability.process_manifest`` weren't being applied to live
containers, so shelled-out HTTP traffic from terminal/python tools
wasn't actually flowing through Caido.
- **Tool exposure**: add the seven Caido tools (``list_requests``,
``view_request``, ``send_request``, ``repeat_request``,
``scope_rules``, ``list_sitemap``, ``view_sitemap_entry``) to
``_BASE_TOOLS`` in ``agents/factory.py`` like every other sandbox
tool. They were already defined in ``tools/proxy/tools.py``.
- **Healthcheck**: ``entry.py`` now ``await``s
``wait_for_http_ready`` + ``wait_for_tcp_ready`` inline after
``session_manager.create_or_reuse`` returns, before any agent runs.
No more capability state, ``configure_host_ports`` plumbing, or
``on_agent_start`` await-the-task indirection.
- **Instructions block**: dropped. The seven proxy tools' docstrings
cover the HTTPQL syntax and usage already; the duplicate prompt
fragment was overhead.
Cascade cleanups:
- Drop ``caido_capability`` from the agent context (was passed to
every ``make_agent_context`` call but only used by the now-deleted
``on_agent_start`` await).
- Strip the capability await branch from
``StrixOrchestrationHooks.on_agent_start``; that hook now does only
the ``tracer.agents`` mirroring it always should have.
- Drop the ``capability`` key from the session bundle.
- Drop ``strix/sandbox/caido_capability.py`` — entire file (207 LoC).
- Drop the per-file ruff ignore for the deleted file.
mypy clean on every touched file. Net -217 LoC.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Tools:
- Add a single ``dump_tool_result`` helper in ``tools/_decorator.py``
and remove the eight identical ``_dump`` definitions from
``proxy/tools.py``, ``file_edit/tools.py``, ``python/tool.py``,
``terminal/tool.py``, ``todo/tools.py``, ``browser/tool.py``,
``notes/tools.py``, ``agents_graph/tools.py``. Imports trimmed.
Net -50 LoC across the tool modules.
run_config_factory:
- Inline the four retry-policy plumbing pieces
(``_RETRYABLE_HTTP_STATUSES``, ``_DEFAULT_MAX_RETRIES``,
``_DEFAULT_BACKOFF``, ``_default_retry_policy()``) into a single
module-level ``_DEFAULT_RETRY`` ``ModelRetrySettings`` literal. The
inputs were never overridden and the helper had one caller.
Tests:
- Drop migration scars from ``tests/test_run_config_factory.py``
(``Phase 1`` / ``C1`` / ``C11`` / ``C21`` / ``HARNESS_WIKI`` / ``AUDIT``
references). Replace the ``_RETRYABLE_HTTP_STATUSES``-touching test
with a ``retry.policy is not None`` smoke check now that the constant
has been inlined.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Tracer:
- Collapse the ``live`` / ``completed`` LLM stat buckets into one
flat dict. The ``completed`` bucket was only ever written by tests
— production never moved stats across, and ``get_total_llm_stats``
always summed both for display.
- Drop ``record_llm_usage(agent_id=...)``: argument was unused, and
the per-call ``bucket=`` knob is gone with the buckets.
run_config_factory:
- Drop unused ``parallel_tool_calls``, ``tool_choice`` parameters
from ``make_run_config`` — no caller ever overrode them.
- Drop ``agent_name`` from ``make_agent_context`` — set into the
context dict but no consumer ever read it; the bus's ``names`` map
is the source of truth.
Wire reasoning_effort through:
- ``Config.get("strix_reasoning_effort")`` is now actually plumbed
to ``make_run_config`` from ``entry.py``. Previously the env var
was advertised but never consumed.
Multi-agent graph tools:
- Replace six copies of
``inner = ctx.context if isinstance(ctx.context, dict) else {}``
with a single ``_ctx(ctx)`` helper.
Todo tools:
- Lift the duplicated ``priority_order`` / ``status_order`` dicts
to module-level ``_PRIORITY_RANK`` / ``_STATUS_RANK`` and replace
both inline sort lambdas with ``_todo_sort_key``.
Notes tools:
- Delete ``append_note_content`` (and its test): docstring claimed
it was for an "agents-graph wiki-update hook on agent_finish" that
was never wired up. Pure dead public API.
Style:
- Drop the ``del ctx`` no-ops from notes / reporting / web_search
tools. ``ARG001`` is already silenced project-wide for tool
modules; the ``del`` was cargo-culted.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Argument parser:
- Delete ``strix/tools/argument_parser.py`` and its tests. The SDK
validates and types tool arguments via Pydantic before they hit our
wrappers, and the in-container tool server receives JSON-typed
kwargs over the wire. The string-coercion belt-and-suspenders is no
longer pulling its weight.
XML → JSON / typed structures:
- ``create_vulnerability_report``: ``cvss_breakdown`` is now a
``dict[str, str]`` of the 8 metrics; ``code_locations`` is a
``list[dict]``. No more XML parsing in the tool or the renderer.
- ``check_duplicate``: the dedup judge now emits a single JSON object
instead of an ``<dedupe_result>`` block. Strict JSON parser handles
optional code-fence wrappers.
- ``agent_finish``: completion report posted to the parent inbox is a
JSON object (``kind``, ``from``, ``agent_id``, ``success``,
``summary``, ``findings``, ``recommendations``) rather than a
hand-rolled ``<agent_completion_report>`` XML envelope.
- ``create_agent``: identity preamble + inherited-context markers are
plain bracketed labels rather than ``<agent_delegation>`` /
``<inherited_context_from_parent>`` envelopes.
- ``inject_messages_filter``: peer messages get a
``[Message from agent <id> | type=... | priority=...]`` header line
instead of an ``<inter_agent_message>`` envelope.
- Crash + system-warning messages: bracketed labels, no XML.
- System prompt: the inter-agent block now describes the new header
format and drops the "never echo XML envelope" rule.
- ``strix/llm/utils.py``: deleted. ``clean_content`` collapsed into a
one-line blank-line normalizer in the agent-message renderer (the
XML envelope scrub had nothing left to scrub).
Tests updated to match the new shapes.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Port the prose guidance that previously lived in the deleted
*_actions_schema.xml files into per-tool docstrings, so the SDK's
auto-generated function schema carries the same domain knowledge
(HTTPQL syntax, Caido sitemap kinds, browser persistence/JS rules,
agent specialization caps, customer-facing report rules, CVSS/CWE
guidance, etc.) without any custom prompt scaffolding.
Strip the <tool_usage> block from system_prompt.jinja — XML format
guidance, the "CRITICAL RULES" 0-8 list, and the </function>
closing-tag reminder all contradicted the SDK's native JSON
function-calling protocol.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Cleanup pass after the migration:
#1 Inline ``*_actions.py`` into wrapper ``tool[s].py`` for the
non-sandbox tools (think, todo, notes, reporting, web_search,
finish_scan). One file per tool family now. Helpers + public
function bodies live alongside the ``@strix_tool``-decorated
wrappers that call them.
For notes, the sync helpers are renamed to ``_create_note_impl`` /
``_list_notes_impl`` / etc. so the public names ``create_note`` /
``list_notes`` / etc. can be the FunctionTool instances the agent
factory imports. ``append_note_content`` (used by the agents-graph
wiki-update hook) calls the impl helpers directly.
#2 Delete ``strix/tools/_state_adapter.py``. The ``AgentStateAdapter``
shim only existed to feed legacy ``*_actions.py`` functions a
``state.agent_id`` they could read. With the actions inlined, the
wrappers read ``ctx.context['agent_id']`` directly.
#3 Strip ``strix/tools/registry.py`` from ~250 LOC to ~110.
Deleted: XML schema loading, ``_parse_param_schema``,
``get_tools_prompt``, ``get_tool_param_schema``, ``needs_agent_state``,
``should_execute_in_sandbox``, ``validate_tool_availability`` — all
for the host-side legacy dispatcher path. Kept the ``register_tool``
decorator (sandbox side), ``get_tool_by_name``, ``get_tool_names``,
``tools`` list, ``clear_registry``.
The Jinja prompt template's ``{{ get_tools_prompt() }}`` injection
is dropped — the SDK auto-generates tool descriptions from function
signatures, so the legacy XML tool block was redundant and stale.
#4 Delete every ``*_actions_schema.xml`` (12 files). They were read
by the now-removed ``_load_xml_schema`` to build the legacy prompt's
tool descriptions. No consumer remains.
Side fixes:
- ``reporting_renderer.py`` updated to import ``_parse_*_xml`` from
the new location with leading underscore.
- ``test_local_tools.py``, ``test_notes_jsonl_concurrency.py``,
``test_notes_wiki.py`` updated to point at the new module paths
and call the ``_*_impl`` sync helpers.
Tests: 279/279 passing. ~1500 LOC of action files moved into the
tool wrappers; ~140 LOC of registry boilerplate removed; ~400 lines
of dead XML deleted.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Critical fixes:
- ``StrixOrchestrationHooks.on_agent_start`` now finds the
``CaidoCapability`` via ``ctx.context['caido_capability']`` instead
of ``agent.capabilities`` (we use plain ``Agent``, not
``SandboxAgent``, so the latter never existed). The session
manager's bundle already exposes the capability; ``run_strix_scan``
threads it through ``make_agent_context`` and ``create_agent``
forwards it to children.
- ``run_strix_scan`` registers the ``StrixTracingProcessor`` with the
SDK's tracing provider via ``add_trace_processor`` so SDK trace
spans hit ``run_dir/events.jsonl`` (was previously a parallel stream
the SDK ignored).
- ``on_llm_end`` now writes to ``Tracer.record_llm_usage`` in
addition to ``bus.record_usage`` so the CLI/TUI stats panel sees
real numbers instead of zeros.
- ``run_strix_scan`` accepts an externally-built ``AgentMessageBus``
+ an explicit ``model`` arg. The TUI pre-creates the bus so its
stop and chat-input handlers can submit ``bus.send`` /
``bus.cancel_descendants`` coroutines onto the scan thread's loop
via ``asyncio.run_coroutine_threadsafe`` — replacing the
TODO-stub no-ops.
- ``model`` config now propagates root → context → child agents in
``create_agent`` (was hardcoded fallback).
Dead-code removal:
- Deleted the ``load_skill`` tool entirely (host module, sandbox
module, TUI renderer, tests). The legacy implementation reached
into a global ``_agent_instances`` registry that no longer exists;
the post-migration stub returned ``success=True`` without
injecting anything — pure theater. Skills are still preloaded via
the system prompt at scan-bring-up.
- Dropped ``tenacity`` and ``xmltodict`` from
``[project.dependencies]`` — neither is imported anywhere
post-migration.
- Stripped the system prompt's "use the load_skill tool" lines.
Tests: 278/278 passing. Removed two ``load_skill`` test cases and a
``test_tool_registration_modes::test_load_skill_import_...`` assertion
that exercised the deleted module.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
0xallam