A manager organizes and validates work, so their own campaigns skip the
draft -> submit -> pending_approval queue and go straight to active with
the start_date they provide (they're the same role that would otherwise
approve it). Manager can also now create tests from the catalog, same as
red_lead/blue_lead.
Leads get a Create Template button that adds directly to the catalog
(existing POST /test-templates). Operators (red_tech/blue_tech) get the
same button, but their proposal now lands in a new template_suggestions
review queue instead — a lead on their team can approve it as-is, edit
fields before approving, or discard it. Modeled on the existing
ProcedureSuggestion review workflow.
- Approve endpoint now only creates Jira tickets immediately when
start_date is now/past; a new periodic job (every 15 min) catches
campaigns whose scheduled start_date has since arrived.
- Recurring campaign clones now go to pending_approval instead of
active, routing through the same manager-approval gate as any other
campaign; managers are notified instead of red_tech.
- Fix UTC conversion for the campaign approval start_date input and
extract shared isoToDatetimeLocal/datetimeLocalToIso helpers.
Manager could not see the Review Queue nav item, load the page, or
call mark-reviewed (red_lead/blue_lead/admin only) — added to the
route guard, sidebar visibility, badge query, and the backend
endpoint's role check.
Webhook Configuration was shown to red_lead/blue_lead in Settings
even though the backend already restricted every webhook endpoint to
admin only — the tab is now admin-only in the UI too, matching what
was already enforced server-side.
Test Catalog's pagination showed only 'Page N' with no way to know
how many pages existed, and the campaign 'Add Test' modal's template
picker had no pagination at all — with 2800+ templates in the
catalog, its flat 100-row fetch made most templates unreachable
unless a search happened to match. Adds GET /test-templates/count
(open to any authenticated user, mirrors the list endpoint's filters)
and wires real 'Page N of M' + total-available counts into both.
Request Discussion only ever nudged the peer lead, and the one-time
manager notification when a dispute starts had no real follow-up
action attached to it — a manager could be told a dispute existed but
had no way to actually do anything about it. Adds:
- POST /tests/{id}/escalate-to-manager — either lead can explicitly
ask managers to step in, distinct from the passive initial notice.
- POST /tests/{id}/manager-resolve-dispute — a manager decides the
final outcome (validated/rejected) directly, bypassing both leads'
votes entirely, reusing the same dual-validation event dispatch so
Jira/notifications behave like any other resolution.
Both leads are notified of the manager's final call, win or lose.
Leads can now edit Expected Detection in the create-from-template
form, same as Suggested Attack Procedure — the edit seeds the new
test's detect_procedure via a detect_procedure_override, without
touching the template itself. The template only updates later through
the existing procedure-suggestion approval flow, once a round is
actually submitted — same mechanism as the red side, no new bypass.
Two overlapping TestTemplate fields (expected_detection: narrative
guidance from imports/leads; detect_suggested_procedure: concrete
commands from approved suggestions) are now one. Blue-side procedure
suggestions target expected_detection directly, appending onto
whatever is already there rather than overwriting it — same
merge-not-overwrite behavior already used for the red side. Existing
detect_suggested_procedure data is folded into expected_detection
before the column is dropped.
canHold checked role OR role instead of matching the current phase, so
a red_tech still saw (and could call) Hold/Resume on a test already
sitting in blue_evaluating. Fixed on both frontend and backend — the
API had the same gap, not just the button visibility.
When Red or Blue submits a round for a test created from a template,
any extractable command in their procedure text is proposed as an
update to the template's suggested procedure — never written
automatically. Adds GET/approve/reject endpoints scoped so a lead only
reviews their own team's suggestions; approving writes the suggested
text into the template, rejecting leaves it untouched.
Red and Blue could not see each other's real field data until both
reviews passed, showing a placeholder message instead. This kept
detection testing blind to what Red actually did; both sides now see
the full, live test data at all times.
- reopen_red_review starts the timer paused (paused_at set) instead of
immediately running, since the operator hasn't resumed working yet —
blue already did this via blue_work_started_at, red needed the same
behavior via the existing pause/resume mechanism.
- update_test_red/update_test_blue now lock edits to the actual assigned
operator for leads too, not just techs — a lead could previously edit
another operator's in-progress test. Mirrored in TeamTabs.tsx.
- push_test_event reassigns the Jira ticket to the original operator
(not the reopening lead) when a test returns to red_executing or
blue_evaluating for rework, instead of leaving it on the lead.
Admin still bypassed require_any_role (non-strict) on the pure
operator actions: start-execution, submit-red, start-blue-work,
submit-blue, pause/resume-timer, and the red/blue field-edit + general
test create/update/remediation/import-rt/template endpoints all used
the loose dependency even where admin wasn't in the role tuple.
Switched every one of these to require_any_role_strict, matching the
review/validate/dispute/hold/assign endpoints already fixed earlier.
Frontend: removed every remaining role === "admin" disjunct gating
Start Execution, Submit to Blue Team, blue evaluating actions, timer
control, red/blue field editing, test creation, and template creation.
Team-blindness visibility (isBlind) deliberately keeps the admin
bypass — admin still sees both sides unmasked for oversight, since
that's visibility, not operating.
Updated ~20 tests whose fixtures used the admin account as a
convenience shortcut to create/drive tests through the workflow —
switched them to a red_lead account, fixing an incidental
fixture-resolution-order cookie bug along the way (client's cookie
jar prefers the last-logged-in role's cookie over any Bearer header
explicitly passed to a later request).
Leads get the same two-queue view operators already have: 'Available
to Review' (pending reviews currently assigned to a peer lead) and
'My Assigned Reviews' (assigned to me), replacing the old single
'My Reviews' toggle. Since the load-balanced auto-assignment always
picks a reviewer immediately, 'available' means peer-assigned reviews
a lead could pick up if the assignee can't get to them, not unclaimed
ones (there aren't any).
POST /tests/{id}/assign now also accepts red_reviewer_assignee /
blue_reviewer_assignee, validated against the matching lead role and
synced to Jira the same way operator assignment already is. The
AssigneeControl UI gained a 'reviewer' kind (leads-only picker) shown
on the test detail header while a test sits in red_review/blue_review
— this also fixes the reviewer assignment being invisible in Aegis
even though it was already being pushed to Jira correctly.
- Fix the red/blue draft form losing unsaved fields (e.g. execution_start_time) whenever an evidence upload refetches the test — the hydration effect now runs once per test, not on every refetch.
- Add color-scheme: dark so native date/time picker popups match the app theme instead of rendering white.
- Fix _STATE_TO_JIRA_STATUS: real Jira statuses are 'To Do' and 'Red Team test review', not 'To-Do' and 'RT Test Review' — confirmed live against the actual workflow transitions, which is why Submit to Blue Team never moved the ticket past 'In Progress'.
- Tempo worklog sync was silently blocked by TEMPO_ENABLED defaulting to False with no admin-facing toggle, even after configuring a Tempo token via Settings. Now bypassed once an admin or personal token is actually configured, mirroring Jira's DB-backed enabled flag.
- Hold now actually pauses the running phase timer (paused_at), and Resume accumulates the held duration into red/blue_paused_seconds — previously the timer kept counting through a hold.
Admin can no longer be picked as a red_tech/blue_tech assignee — it
administers the site, it doesn't operate tests. Applied to the
eligible-assignee list on both the picker UI and the assign endpoint's
validation, and dropped from the GET /users/operators pool.
push_assignee_update() now falls back to a fresh Jira account-id lookup
when the assignee hasn't logged in yet (so jira_account_id is still
empty), instead of silently no-op'ing — assignment now syncs to Jira
immediately regardless of whether the assignee has ever logged into
Aegis before.
Admin administers the site — it no longer has any override on
validate-red/blue, review-red/blue, resolve-dispute, request-discussion,
reopen, hold, resume, assign-operators, or classification-update.
Introduces require_any_role_strict(), a role dependency without the
global admin bypass, so these specific endpoints truly exclude admin
instead of only removing it from the (redundant) role tuple.
Managers gain the ability to assign red_tech/blue_tech operators
(POST /tests/{id}/assign, GET /users/operators) alongside leads, since
that's coordination, not resolving a ticket.
Also enlarges and repositions the operator-assignment controls next
to the Start Execution button, and fixes a literal '\u2014' rendering
as text instead of an em dash in the test detail technique line.
GET /system/jira-config was admin-only, so non-admin users hit a 403
and the frontend silently fell back to a hardcoded jira.atlassian.com
base URL when building /browse/{key} links. Opened it to any
authenticated user (still no secrets returned).
Also drops the redundant 'security-test' Jira label, and adds a
page-size selector (10/25/50/100) and a 'no existing test yet' filter
to the Test Catalog.
Adds GET /users/operators (leads+admin) and wires a lead-only assign
control into the test detail header, calling the existing but
previously unreachable POST /tests/{id}/assign endpoint. Manual
assignment now also pushes the Jira ticket assignee immediately
instead of waiting for the operator to start execution.
Also adds test.platform as a kebab-case label on Jira ticket creation.
GET /tests caps limit at 200, so any page relying on len(results) for a
"total" count (e.g. Validated Tests) silently under-reports once there
are more matches than the page size — the 200 shown was a page cap, not
the real total.
- Adds GET /tests/count, sharing its query-building with list_tests()
via a new _build_test_query() so the two can't drift apart.
- Adds technique_search (free-text on MITRE ID or name), attack_success,
detection_result, and validated_from/validated_to filters, so callers
aren't limited to state/technique_id/platform/creator.
POST /system/jira-test bundled 4 independent checks (enabled, url,
admin_email, admin_api_token) behind has_admin_jira_configured() but
always blamed "Admin Email and Admin API Token" regardless of which one
failed. Found via live QA: URL/email/token were all correctly saved, but
the "Enable Jira Integration" toggle was never switched on — the error
message pointed at the wrong fields, making it look like the already-set
credentials were missing.
_REDACTED_KEYS listed jira.api_token/jira.password/tempo.api_token, but
routers/system.py actually writes admin credentials under
jira.admin_api_token and tempo.admin_token. The mismatch meant every
config export bundle included the live Jira and Tempo admin API tokens
in plaintext instead of redacting them. Found while researching the
Jira/Tempo integration for a service-account request; added a
regression test.
- start_execution now blocks a test from starting before its campaign's
start_date, closing a gap where the field was persisted but never
enforced
- GET /tests gains a reviewer_id "my reviews" filter for the red_review/
blue_review lead-gate stage, distinct from pending_validation_side
(which only ever covered the later in_review stage and ignored
assignment entirely)
- get_coverage_by_tactic now returns all 14 MITRE tactics in canonical
kill-chain order instead of an alphabetical, partial list — the
regular Dashboard and Executive Dashboard both consume this endpoint
and previously disagreed on order/completeness
- test-template listing now includes existing_test_count per technique
so the catalog can warn before creating a likely-duplicate test
- /auth/refresh now allows a short grace window past expiry and checks
the blacklist, so an active session's silent refresh no longer fails
the instant its own token expires
- normal manager /approve flow now creates Jira tickets for the campaign
and its already-linked tests, matching the admin-only /activate path
- GenerateFromActorPayload now accepts start_date and threads it through
to the new campaign instead of silently discarding it
Now that only the manager sets start_date, and only at the moment of
approval (which immediately activates the campaign), a draft campaign
can never have a start_date. This made three things permanently
unreachable: the hourly _run_scheduled_campaign_activation cron job,
the activate_campaign 409/force future-date guard, and a query filter
hiding tests from scheduled-but-inactive draft campaigns. Removes all
three rather than leaving dead code behind. Also adds the missing
manager-cannot-directly-activate router test.
Moves the local _post cookie-clearing helper into a shared 'api' fixture
in conftest.py so later router tests in this plan (Tasks 10/11) don't
have to reinvent or forget the TestClient cookie-vs-Authorization-header
gotcha. Also adds a one-line comment at both require_any_role("manager")
call sites clarifying admin passthrough is automatic.
- Single admin Jira account stored in system_configs (jira.admin_email, jira.admin_api_token)
- Admin Tempo token in system_configs (tempo.admin_token)
- get_admin_jira_client() + has_admin_jira_configured() replace per-user auth in all lifecycle hooks
- lookup_user_jira_account_id() auto-discovers each user's Atlassian accountId by email on login
- auto_log_test_worklog() now uses admin Tempo token and logs both red+blue team time
- Settings > Jira: admin credentials fields + test buttons moved to admin Jira tab
- Profile section simplified: jira_account_id shown read-only (auto-detected)
Tar Slip (CWE-22) — 3 import services:
threat_actor, lolbas, caldera: add path validation before extractall()
to prevent malicious zip members with ../ escaping the target directory.
(sigma, elastic, atomic already had this protection)
Path Traversal (CWE-23) — professional_reports.py:
Add _assert_safe_report_path() check on all 5 report endpoints to
verify the generated filepath stays within REPORT_OUTPUT_DIR.
Open Redirect (CWE-601) — sso.py:
Validate IdP redirect URL scheme (must be http/https) before
issuing RedirectResponse, blocking javascript: and data: redirects.
DOM XSS (CWE-79) — 4 frontend pages:
Create src/utils/url.ts with safeUrl() that rejects non-http/https
protocols; apply to actor.mitre_url, ref.url, intel.url.
Sanitize framework name to alphanumeric-only before DOM insertion.
Restrict evidence MIME types to an explicit safe allowlist (png/jpg/gif/webp).
Hardcoded credentials (CWE-798):
verify_gaps.py, create_wiki.py: replace literal passwords with
environment variable reads (AEGIS_ADMIN_PASSWORD, GITEA_PASSWORD).
- config.py: move REPORT_OUTPUT_DIR from /tmp (world-writable) to /app/reports
to prevent CWE-377 symlink attack vector (B108, only real security issue)
- main.py: log startup seed failures instead of silently swallowing them (B110)
- Add # nosec annotations to intentional try/except patterns that are by design:
Jira integration errors, email failures, DetachedInstanceError, storage errors,
and Jira session timeout (all B110/B112 false positives)
- Add # nosec B105 to false positives where bandit misidentifies config key
names and masking strings as hardcoded passwords
- Add .bandit config to skip B311 in seed_demo.py (random used for fake
demo data generation, not cryptographic purposes)
Snyk scan found 3 High severity vulns: two in ecdsa (pulled by python-jose)
and one in diskcache (pulled by pySigma, never imported). Remove both
vulnerable dependencies and migrate JWT handling to PyJWT. Fix
test_logout_revokes_token which broke because test stubs sys.modules[jose]
with a MagicMock at collection time; test now uses PyJWT directly.
Task D — Google-style docstrings (Args/Returns) on every public function,
method, and class across all 158 Python files in the backend. Zero ruff D
violations (pydocstyle Google convention).
Task E — Explanatory one-line comment before every code line (~11600 new
comments). ruff check passes clean after isort re-sort.
kitos