blitzstrike 1.1.5 → 1.2.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/CHANGELOG.md +31 -0
- package/dist/index.js +1687 -1671
- package/package.json +1 -1
- package/skills/bs-orchestrate-engagement.md +12 -2
package/CHANGELOG.md
CHANGED
|
@@ -5,6 +5,37 @@ All notable changes to Blitz Strike are documented here.
|
|
|
5
5
|
The format follows [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
|
|
6
6
|
and this project adheres to [Semantic Versioning](https://semver.org/).
|
|
7
7
|
|
|
8
|
+
## [1.2.0] — 2026-09-13
|
|
9
|
+
|
|
10
|
+
### Added — deterministic full live pipeline (no-babysitting)
|
|
11
|
+
|
|
12
|
+
- `run_engagement` on a URL target now runs the FULL autonomous live pipeline in
|
|
13
|
+
one deterministic call: `live_recon` (fingerprint, WAF, tech, crawl,
|
|
14
|
+
subdomains, DNS, API discovery, intel, next_steps) + marker/negative-control
|
|
15
|
+
reflected-input sweep + inline report. The LLM no longer drives recon →
|
|
16
|
+
analyze → verify step-by-step; a single call maps the whole surface and
|
|
17
|
+
verifies reflections. The server-side orchestrator (not the LLM) drives the
|
|
18
|
+
flow, so it cannot stall to ask the user.
|
|
19
|
+
|
|
20
|
+
## [1.1.7] — 2026-09-13
|
|
21
|
+
|
|
22
|
+
### Improved — no-babysitting doctrine refined from OMOP tricks
|
|
23
|
+
|
|
24
|
+
- Pause conditions now precise (3): destructive/DoS needs consent, external
|
|
25
|
+
dependency beyond control blocks (e.g. phone OTP), or engagement complete.
|
|
26
|
+
- Added "no retry cap" (diagnose + resume until verified, never move on
|
|
27
|
+
unverified) and "parallel by default" (fire all independent workstreams in
|
|
28
|
+
one message; only named input/file-conflict dependencies are sequential).
|
|
29
|
+
|
|
30
|
+
## [1.1.6] — 2026-09-13
|
|
31
|
+
|
|
32
|
+
### Fixed — never ask "should I continue?" (no-babysitting)
|
|
33
|
+
|
|
34
|
+
- Doctrine now explicit: NEVER ask "should I continue?" / "do you want me to
|
|
35
|
+
stop?" — continue autonomously to the end. Surface to the user ONLY when a
|
|
36
|
+
genuinely destructive/DoS action needs consent, or the engagement is complete
|
|
37
|
+
(final report). Removed "target is ambiguous" as a mid-engagement pause reason.
|
|
38
|
+
|
|
8
39
|
## [1.1.5] — 2026-09-13
|
|
9
40
|
|
|
10
41
|
### Expanded — full Awesome-Hacking resource index (78 entries, 44 domains)
|