The wake lines interpolated a planner-written intent summary into the trusted sentence, and the overview kept every summary in the plain JSON, so an inherited summary crossed tasks as if this planner had written it (#124, F3). Frame the summary on the wake lines and move the open/running/recent-done summaries into one framed block; the structured intent fields stay plain.
A task already inside the wrap-up reserve only has that reserve left. The worker estimate starts from the defaults because usage rows have no run boundary.
Planner, worker, judge, and compactor rows rebuild the cap. Main-agent chat, goal decomposition, and side questions stay out, so a restart does not count turns the running process never metered.
The recording proxy now stores WebSocket and Server-Sent Events streams: one exchange per stream (101 websocket, or the real text/event-stream response) and its messages in stream_messages, with per-message and per-exchange limits and blob spill. Messages are readable through Traffic.StreamMessages, GET /api/traffic/stream and the traffic_messages agent tool; deletion and blob GC cover them.
The NOA budget refusal now wraps an exported ErrNoaBudget sentinel, which survives the Norma harness unchanged; retryableWorkerModelError excludes it, so a deterministic refusal is no longer replayed twice by the engine.
Bump Norma to v0.4.9 (BaaRai/Norma v0.4.9). enableNoa passes the configured context window to noaadapter as ModelContextLimit and, once NOA is enabled, wraps the provider in a budget check that refuses a request whose estimate exceeds that window before any transport.
Resume errors from a corrupt or unreadable transcript are returned as "restore Norma session %q: %w" before any model or tool call instead of being dropped; a missing transcript still starts a fresh session. The session id is set and put on the context regardless of the transcript store, so the provider and usage metering see the same identity.
Mechanical: gofmt -w on the 11 files gofmt -l listed on main. Besides spacing and alignment, gofmt sorted the imports inside one group in two test files and removed a trailing blank line in a third. The go job now fails, listing the files, when gofmt -l reports any tracked Go file.
The low-and-slow block said that commands launched through Bash are not paced. With traffic capture on, Bash commands that use the capture proxy are now paced by rate and jitter; raw scanners and commands that bypass the proxy are not, so the rate flags remain the brake.
With traffic capture on, the Bash proxy URL carries a marker userinfo. The recording proxy drops Proxy-Authorization from every request before it is forwarded or recorded; when the value is the marker it marks the client connection, and every later request on it, including those inside its CONNECT tunnel, waits for the per-host token and jitter of the current noise profile (egress.Wait) and is counted, with a bounded wait. WebFetch, the http custom tools and enrich keep the unmarked URL, which egress already paces, so nothing is paced twice. Off paces nothing. Best effort: only clients that honour the proxy are covered, with capture on.
The egress governor is now a current governor that SetProfile swaps: new requests use the new one, requests in flight finish on the old one, and one package-level ticker logs the counters of the current governor. WebFetch, the http custom tools and the enrich probe resolve it per request, and the agent prompt block reads the current profile. The noise_profile setting (off, quiet, balanced, aggressive) is applied at startup and wins over ARTEX_NOISE_PROFILE, which stays the fallback; PUT /api/settings validates, stores and applies it without rebuilding agents, and GET /api/settings/noise returns the per-host counters without resetting them. System settings gets a Target noise card.
PromptOverride, WrapupOverride, WrapupMaxTurnsOverride,
WrapupTaskTimeoutOverride and WrapupTaskTimeoutTurnsOverride are wired
in server.New while restored tasks may already render prompts and build
settlement configs on other goroutines. go test -race ./server reported
races between server.New (server.go:195/207/215) and agent.renderSystem /
resolveWrapup / resolveWrapupTurns.
Same treatment as the tool hooks: unexported variables behind a
sync.RWMutex (one for the prompt hook, one shared by the four wrap-up
hooks), Set* setters and getters, and readers that copy the function
value under the lock and call it outside. A nil hook still means the
built-in default.
TestAgentHooksConcurrentSwap now covers all eight hooks.
server wiring (wireAgentAugment, wireTools) assigned the package-level
hooks agent.ToolAugment and agent.ToolResolve while restored tasks could
already be running AugmentTools on other goroutines. go test -race ./server
reported this as a DATA RACE between agent.AugmentTools and
server.wireAgentAugment.
Each hook is now an unexported variable behind a sync.RWMutex, with a
setter (SetToolAugment, SetToolResolve) and a getter (ToolAugment(),
ToolResolve()). AugmentTools copies the function value through the getter
and calls it outside the lock. A nil hook still means no augmentation /
no resolution, as before.
FindingTrafficBindingEnabled is wired in the same server/assembly.go code
and read on the same AugmentTools path, so it gets the same treatment.
Adds TestAugmentToolsHooksConcurrentSwap, which runs AugmentTools while
the setters swap the hooks, so -race catches an unsynchronised access
without depending on test ordering.
- describe the graph stores as PostgreSQL in the entrypoint and agent package
- document that -data holds the traffic index, task files and transcripts
- add an Unreleased note for the untrusted-output boundary default flip
- use a placeholder version in the README/build.sh release examples
scanURL kept the userinfo of httpx, ffuf and nuclei URLs, so a line such
as http://user:pass@host/path stored the credentials in assets.url, in
the notable items and in the facts. scanURL now drops u.User before
re-serializing; scheme, host, port, path and query are kept.
The nuclei fact detail also embedded the raw host and matched-at values,
the httpx detail the raw host and the ffuf detail the raw redirect
location. They now go through stripUserinfo, which removes only the
userinfo of a URL or bare host:port and keeps every other byte, so a
value without userinfo is unchanged and fact deduplication is the same.
bbotTyped drops the userinfo of the typed url of an in-scope URL event
before it becomes an asset.
Rows already stored with userinfo are not rewritten here.
Extends ingest_scan_output with three formats:
- nuclei-jsonl: one tool_output fact per result and an asset for the
scanned target when it is valid; no automatic finding
- httpx-jsonl: one HTTP service asset and one fact per URL
- ffuf-json: one URL asset and one fact per result, read token by token
with encoding/json; a file cut short keeps the complete results
JSONL lines go through readJSONLines, taken from the bbot ingester
(64 KiB per line; longer lines and invalid JSON are skipped and
counted). bbot now uses the same reader, with no behaviour change.
One file yields at most 50000 assets and facts; records past the cap
are counted in over_cap and the result is marked truncated.
A worker can write nmap output with -oX and call ingest_scan_output with
the format and path. The file is parsed deterministically: assets go
through the insert_assets handler, each host becomes one tool_output
fact listing its open ports, and the model gets only counters and at
most 20 notable items per type.
- path is resolved under the run working directory; "..", an absolute
path outside it and a symlink leading outside are refused; 256 MiB cap
- the XML is streamed host by host with encoding/xml; a file cut short
keeps the complete hosts and returns truncated: true
- target text (banners, products, hostnames) is kept only as a
Go-quoted string in the fact detail
- re-ingesting a file upserts the same assets and skips identical facts
Worker-only, like ingest_bbot. nuclei, httpx and ffuf follow separately.
Dalfox v3.2.4 (Rust) has no -w and exposes a global --rate-limit plus --workers (docs/content/reference/cli.md). Subfinder -t only bounds concurrent DNS lookups in active mode, so the passive line is removed; the generic fallback covers it.
When ARTEX_NOISE_PROFILE is set, the planner, worker, main agent and chat system prompts gain a code-owned block with the profile per-host limits and exact rate/delay flags for the scanners installed in the image. With no profile the block is empty, so every prompt is byte-for-byte unchanged.
README documents the block.
The LLM fallback judge gets a closed mode (default when the judge is on):
the model answers eight code-owned questions with yes|no|unknown and one
evidence line each, in one call; the decision is taken in code. Every
failure (invalid output, missing key, value outside the vocabulary, a
failed call, arguments above the 256 KiB cap) is ask. The custom judge
prompt becomes bounded operator-policy data (8000 runes). The answers are
stored in the intercept audit and shown in the approval detail.
freeform keeps the previous behaviour; defaultJudgeFailAction changes
from allow to ask.
Normalise and length-check ledger fields in db.RecordCoverage, frame the planner gap lines in the data envelope, cap the wrap-up (25) and report (200) gap lists, cap current rows at 2000 per task. Replace the activity budget marker with tasks.end_reason, written with the guarded terminal status and cleared on revive. Generic 500 message in taskCoverage; early return on an empty ledger in the planner block.
Workers record per (surface, risk_area) outcomes with the new record_coverage
tool (reported, no_issue_found, ruled_out, not_applicable, needs_follow_up).
Gaps are needs_follow_up entries, entries missing their required evidence or
reason, and goal vulnclass values with no entry (unrecorded_risk_class), so an
absent finding never reads as a clean target. Completeness marks timeout,
budget wrap-up, pause, abort and running tasks as incomplete with a caveat.
Surfaced in the planner context (one line per round, full gap list on the
timeout/budget wrap-up round), GET /api/tasks/{id}/coverage, the coverage card
and the Markdown report. Follows coverage_enabled; no new env var.
- resetSettledRun clears settling and the budget ledger; a done/failed task
revived to running now plans and claims again (was gated by the old wrap-up).
resetTimeoutRevival calls it. Test: TestResetSettledRunRestartsBudget.
- docker-compose.yml passes the four ARTEX_*_BUDGET_* variables through.
- .env.example and README document the variables, the wrap-up, the in-memory
ledger and the practical floor.
- llmrec: the usage sink runs even without a PostgreSQL recorder; comment fixed.
- BudgetWrapupBlocks allows update_finding_report.
Per-level token/turn budgets: task -> intent (worker run) -> planner round
(counted inside the task). A child's effective budget is
min(own budget, parent's remaining above the wrap-up reserve). When the task
can no longer afford one worker, workers stop claiming intents and the task
settles through one forced wrap-up planner round (the existing settle
sequence, budget words instead of the timeout words).
Config: ARTEX_TASK_BUDGET_TOKENS, ARTEX_TASK_BUDGET_TURNS,
ARTEX_INTENT_BUDGET_TOKENS, ARTEX_INTENT_BUDGET_TURNS. 0 or unset = unlimited;
with all four unset nothing changes.
ARTEX_NOISE_PROFILE = off | quiet | balanced | aggressive drives one shared
governor used by WebFetch, the http custom tools and the enrich probe: token
bucket, bounded jitter and concurrency per host, a global cap, backoff on
429/503 honouring Retry-After with bounded retries, per-task coalescing and
a short cache of identical GETs (never POST/PUT/PATCH/DELETE), and a
per-host noise log line each minute. Off (the default) is a strict
passthrough.
Removes the fixed artex-enrich/1.0 User-Agent; ARTEX_TARGET_USER_AGENT sets
one under a profile (empty = unchanged).
When ARTEX_TOOL_RAG is enabled, AugmentTools keeps the core tools plus the
top-k (ARTEX_TOOL_RAG_TOPK, default 12) BM25-ranked bound tools for the
worker intent / planner goal. Runs after ToolResolve, so it only removes;
falls back to the full list on empty query, catalog under 20, or scorer error.
ARTEX_UNTRUSTED_BOUNDARY unset or empty now enables the boundary.
0, false, off and no (case-insensitive, trimmed) disable it; any other
value keeps it on. Docs state the default, the opt-out and the measured
cost (about 154 bytes per tool result plus one system note).
ingest_bbot reads a bbot output file by path and ingests it through
the typed parser from #199 (parseBbotJSONL), one line at a time:
- Lines are read with a fixed cap (bbotMaxLineBytes = 64 KiB, newline
included). A longer line is skipped and counted, never buffered whole.
- scope_distance == 0 events that are assets (assetItems) go through
the existing insert_assets handler; no asset SQL is written here.
- Every other event is recorded as a fact with source
third_party_intel. The summary is built from typed fields only; the
free text is stored in detail as the inert Go-quoted string.
- Intel never reaches insert_assets or task_scope.
The tool is in WorkerTools, so the existing BuiltinToolSeeds/SeedTool
loop inserts its row (a new key, inserted on fresh and existing DBs).
It is listed in the target gate's noTargetTools: its input is a local
path, not a network target (TestDomainToolsDeclareNoTarget).
Parse bbot output.json reading only the typed fields (type,
scope_distance, host, url, port, module, module_sequence), validated.
The free text (data, or data_json with the FINDING description) is kept
only as one Go-quoted inert string, never used for scope or an action.
Unknown keys are ignored. Events at scope_distance 0 are the only ones
mapped to insert_assets items, from their typed host; any other distance
becomes a third_party_intel item that is never auto-scoped.
doHTTP now connects through intercept.PinnedDialer under the target scope of the request's context, so the probe uses the same pinned-address dial as the target gate. ProbeSite takes the context of the call that asks for the probe. A probe without a scope is refused; with the gate on, the probe bypasses the recording proxy so the dialer decides on the target itself. The path stays dormant: nothing calls ProbeSite on main.
declaredScopeFilter treats only root_domain, subdomain, ip and cidr rows as
roots. A declared company, icp or keyword row does not make an auto row
count, but the add_task_scope description said "a goal target or scope
added here or manually". The description and the code comments now say
"declared domain or IP root" and name the kinds that act as roots.
Auto task_scope rows (source='auto') come from insert_assets and intent
asset_ids, i.e. from target-derived content. They now count toward the
coverage denominator and list_untested_assets only when they sit under a
root the same task declared (a non-auto row: goal target registered via
add_task_scope, or manual scope): the same or a deeper domain of a declared
root_domain/subdomain, or a net inside a declared ip/cidr.
The direct-source coverage view also stops counting an asset from an intent
anchor alone; fact anchors still count, so tested evidence is unchanged.
Without this, insert_assets put a planted host back into the backlog
through its intent anchor.
task_scope is described as "coverage scope, not enforced" in the code
comments, the add_task_scope description, the goal-decomposer tail and the
task overview UI, instead of "authorization boundary".
Audit F8, 0.7.0.
The planner and worker user messages now carry worker output, finding
details, recent_facts, finding_list, cold_digests, related_tasks and hints
not written by the operator inside a code-owned data envelope whose nonce
comes from crypto/rand (16 bytes, hex) once per message. Embedded marker
stems are neutralised with a fixed-string replacement. The task goal, goal
nodes, constraints and the operator's own hints stay outside, labelled
operator. Overview fact summaries are capped at factSummaryMax (160 runes),
the cap list_facts already uses.
record_fact takes a source field from a closed enum (target_response,
tool_output, third_party_intel, operator_input, inference). An empty value
is stored as inference; any other value is refused with the allowed list.
The source is stored in the fact payload.
The writer of a node is already recorded by the server in the origin
column. compactNode now surfaces origin and source, so recent_facts in the
overview, list_facts, expand_digest members and related-task facts carry
them. The cold-digest compressor input carries source and origin next to
confidence; the digest body is unchanged model text.
prove_goal refuses a fact whose source is target_response, in addition to
the existing confidence=observed rule. The goal stays open; findings are
still accepted.
The worker, pentest and planner default prompts ask for the source, and the
planner treats target_response facts as non-authoritative for goals.
Refs #124 (F4).
Add two optional free-text fields to findings so a reporter can record
what was tested and did not work (counterevidence) and why the
conclusion holds (confidence_rationale).
- db: idempotent ADD COLUMN IF NOT EXISTS, TEXT NOT NULL DEFAULT '',
no CHECK; RecordFindingInput, the INSERT, the shared select list and
all three scan sites carry them.
- agent: report_finding accepts both as optional strings; the worker
and pentest prompts say to fill them when confidence is not high.
- server: FindingDTO (list and detail) carries both, omitted when empty;
existing databases get the two schema properties added to the stored
report_finding row once, without overwriting user edits.
- web: the finding detail page shows each section only when non-empty.
Both fields stay optional: a missing field never refuses a finding.
Audit findings F5, F6 and F11 for issue #124.
F5, goal guards:
- goal_met refuses while any goal node of the task is open, and when the
goals cannot be read. A task without goal nodes keeps the old behavior.
- prove_goal accepts a fact as evidence only when its confidence is
"observed"; otherwise it refuses and the goal stays open. Findings are
still accepted.
F6, hint origin:
- add_hint stores the calling ToolSet's worker as the hint origin instead
of "human", and passes it to the planner trigger (TriggerEvent.HintOrigin).
- renderTriggers names the writer: the human via the main agent, a
cross-task orchestration, or an agent, with a neutral fallback.
F11, built-in rules for the planner and the main agent:
- Interceptor.MatchBuiltin evaluates the enabled "[built-in]" deny rules on
the intercept-enabled tools.
- The target-only guard applies them before the target gate, whether the
gate is on or off. User rules and the judge stay out of it.
- SetRulesForTest is a test seam for the store-less guard tests; Log is a
no-op without a store.