Katana

No entries match these filters. Clear filters

1. v1.8.0 Oct 5, 2026 · issue 095

   Katana v1.8.0 adds recorded-flow headless auth replay, chrome-headless-shell support, and per-page crawl hooks for the hybrid engine.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠

   └──▷ GET THIS VERSION

   copy

   $ git clone --branch v1.8.0 https://github.com/projectdiscovery/katana.git
   # already have the repo? check out this version:
   $ git checkout v1.8.0

   └──▷ TRY IT
   Replay a recorded login flow before crawling an authenticated app — record the login once in Chrome DevTools and pass the exported JSON so katana handles auth automatically.
   copy

   $ katana -u https://app.example.com -headless -rf ./login-flow.json

   • Adds -rf / --recorded-flow flag to replay a Chrome DevTools Recorder JSON (or explicit steps file) before a headless crawl, enabling authenticated crawling without manual session setup.⁠︀︄︇︀︅︉︀︅︀︀︃︀︀︅︅︀︅︊︀︄︆︀︃︄⁠
   • Adds BrowserContextID exposure to the hybrid crawler, allowing callers to correlate crawl traffic to a specific browser context.
   • Adds optional per-page BeforeNavigate and AfterLoad hooks to the hybrid engine, enabling custom logic to run at each page lifecycle event during a crawl.

   3 details in Everything

2. v1.7.0 Aug 5, 2026 · issue 037

   Katana v1.7.0 adds secrets/endpoints extractors to the knowledge base, content similarity deduplication, and headless callbacks.

   └──▷ GET THIS VERSION

   copy

   $ git clone --branch v1.7.0 https://github.com/projectdiscovery/katana.git
   # already have the repo? check out this version:
   $ git checkout v1.7.0

   └──▷ TRY IT
   Crawl a target and extract secrets from the knowledge base, then validate them live against their providers.
   copy

   $ katana -u https://example.com -kb -kb-secrets -kb-validate-secrets

   Crawl a target and skip near-duplicate pages using TF-IDF similarity scoring to reduce noise on large sites.
   copy

   $ katana -u https://example.com -pcs -pcsm tfidf -pcst 0.90

   • Adds -kb-secrets flag to enable a secrets extractor within the knowledge base, automatically detecting exposed credentials and tokens during crawls.
   • Adds -kb-validate-secrets flag to validate detected secrets live against their respective providers (sends real API calls).
   • Adds -kb-endpoints flag to enable an endpoints extractor that classifies REST, GraphQL, SOAP, and XHR requests found during crawling.
   • Adds -pcs / -page-content-similar flag for Layer-2 content similarity filtering to deduplicate near-identical pages during crawls, with configurable mode via -pcsm / -page-content-similar-mode (simhash, tfidf, or bm25; default simhash).
   • Adds -pcsd / -page-content-similar-distance (SimHash max Hamming distance, default 3), -pcst / -page-content-similar-threshold (TF-IDF/BM25BM25The classic keyword-ranking formula behind traditional search engines. Hybrid search combines it with embedding similarity so exact terms and meaning both count. min score 0–1, default 0.85), and -pcsn / -page-content-similar-budget (pages fully processed per similarity cluster, default 1) flags to tune deduplication behaviour.

   +3 more show less

   • Adds -sdd / -similarity-deduplication as a convenience alias for -pcs.
   • Adds callbacks to the headless engine, enabling programmatic interception of headless browser events.
   • Adds a Context field to the Options struct for external lifecycle control — cancelling the provided context.Context stops the rate limiter, crawl session, queue, known-files requests, and headless browser.

   8 details in Everything

3. v1.6.0 May 4, 2026 · release notes

   Katana v1.6.0 adds automatic login, per-domain page budgets, per-host rate limiting, and page load strategy control.

   └──▷ GET THIS VERSION

   copy

   $ git clone --branch v1.6.0 https://github.com/projectdiscovery/katana.git
   # already have the repo? check out this version:
   $ git checkout v1.6.0

   └──▷ TRY IT
   Crawl a JS-heavy SPA that loads content after network activity settles, switching from the default heuristic to networkidle.
   copy

   $ katana -u https://example.com -headless -pls networkidle

   • Adds automatic login support for crawling authenticated targets.
   • Adds per-host rate limiting with adaptive backoff to avoid overwhelming individual hosts.
   • Adds -pls/-page-load-strategy flag to control browser page load timing (heuristic, load, domcontentloaded, networkidle, none).
   • Makes jsluice cgo-free, removing the CGo build dependency for easier cross-compilation.

   4 details in Everything

4. v1.5.0 Mar 10, 2026 · release notes

   Katana v1.5.0 adds URL deduplication, headless captcha solving, and knowledge-base classification

   └──▷ GET THIS VERSION

   copy

   $ git clone --branch v1.5.0 https://github.com/projectdiscovery/katana.git
   # already have the repo? check out this version:
   $ git checkout v1.5.0

   └──▷ TRY IT
   Reduce noise in large crawls by collapsing parameterised or enumerated URL paths into representative samples.
   copy

   $ katana -u https://example.com -filter-similar -o filtered_urls.txt

   • New -filter-similar flag deduplicates crawl output by collapsing URLs with similar path patterns.
   • Adds captcha detection and automatic solving support for headless browser crawling.
   • Adds knowledge-base classification of crawled content using dit.

   3 details in Everything

5. v1.4.0 Jan 6, 2026 · release notes

   Katana v1.4.0 adds headless browser crawling with optional visual display and diagnostic logging.

   └──▷ GET THIS VERSION

   copy

   $ git clone --branch v1.4.0 https://github.com/projectdiscovery/katana.git
   # already have the repo? check out this version:
   $ git checkout v1.4.0

   └──▷ TRY IT
   Crawl a JavaScript-heavy app with headless mode while watching the browser in real time to verify crawl behaviour.
   copy

   $ katana -u https://cloud.projectdiscovery.io -headless -show-browser

   Deep headless crawl of a target, following all form submissions, to maximise attack-surface discovery.
   copy

   $ katana -u https://ginandjuice.shop -headless -depth 5 -aff

   Investigate headless crawl state failures by capturing full diagnostics and verbose logs to a dedicated output directory.
   copy

   $ katana -debug -u https://www.github.com -headless -enable-diagnostics

   • Adds -headless flag to run a full headless browser crawler, enabling JavaScript-rendered page crawling.
   • Adds -show-browser flag to visually display the headless browser during crawling for debugging and inspection.
   • Adds -enable-diagnostics flag to write diagnostic information to a new directory alongside verbose logging for investigating crawl state failures.

   3 details in Everything

6. v1.3.0 Dec 1, 2025 · release notes

   Katana v1.3.0 adds flags to bypass default extension filters and filter JSON output by fields.

   └──▷ GET THIS VERSION

   copy

   $ git clone --branch v1.3.0 https://github.com/projectdiscovery/katana.git
   # already have the repo? check out this version:
   $ git checkout v1.3.0

   └──▷ TRY IT
   Crawl a target without skipping any file extensions — useful when hunting for exposed config, backup, or archive files that the default filter suppresses.
   copy

   $ katana -u https://example.com -no-default-ext-filter -o results.txt

   • New -no-default-ext-filter flag disables Katana's built-in extension filter, allowing crawls to surface URLs that would otherwise be silently dropped.
   • New -eof and -lof flags enable filtering of JSON output by specific fields, giving practitioners tighter control over crawl result ingestion.

   2 details in Everything

7. v1.2.2 Aug 30, 2025 · release notes

   Katana v1.2.2 adds cookie support during crawls and an option to disable the unique content filter.

   └──▷ GET THIS VERSION

   copy

   $ git clone --branch v1.2.2 https://github.com/projectdiscovery/katana.git
   # already have the repo? check out this version:
   $ git checkout v1.2.2

   • Adds cookie support during crawling, enabling authenticated or session-aware crawl workflows.
   • Adds option to disable the unique content filter, allowing duplicate content to be captured during crawls.

   2 details in Everything

8. v1.2.0 Jul 23, 2025 · release notes

   Katana v1.2.0 adds custom output templates, parent-path climbing, and broader XHR/Fetch/Script capture.

   └──▷ GET THIS VERSION

   copy

   $ git clone --branch v1.2.0 https://github.com/projectdiscovery/katana.git
   # already have the repo? check out this version:
   $ git checkout v1.2.0

   └──▷ TRY IT
   Extract emails and their source URLs from a crawl in a pipe-friendly format for downstream triage.
   copy

   $ katana -silent -u https://www.google.com/appserve/security-bugs/m2/new -ot "{{email}} - {{url}}"

   Discover unlisted parent directories of a deep URL — useful when you only have a leaf path and want to enumerate the tree above it.
   copy

   $ katana -silent -u https://scanme.sh/test/path/to/file.html -path-climb

   • New -output-template (-ot) flag lets you define custom output formats using Go-style templates (e.g., {{email}} - {{url}}) to tailor crawl result structure.
   • New -path-climb flag auto-discovers and crawls parent paths from a given URL, useful for uncovering hidden directory structure.
   • Introduces OnSkipURL SDK callback for finer-grained control over URL skipping behavior in library integrations.
   • Network request capturing now includes XHR, Fetch, and Script resource types when extraction mode is enabled, broadening JavaScript-heavy app coverage.
   • JavaScript file endpoint detection now handles a broader range of relative path formats for more accurate matching.

   5 details in Everything

9. v1.1.2 Dec 2, 2024 · release notes

   Katana v1.1.2 adds content-length field in output and makes technology detection opt-in via new -tech-detect flag.

   └──▷ GET THIS VERSION

   copy

   $ git clone --branch v1.1.2 https://github.com/projectdiscovery/katana.git
   # already have the repo? check out this version:
   $ git checkout v1.1.2

   └──▷ TRY IT
   Crawl a target and capture technology stack alongside content lengths — useful for recon pipelines that feed into nuclei or reporting.
   copy

   $ katana -u https://example.com -td -jsonl -o results.jsonl

   • Adds content_len field to crawl output for captured responses.
   • Introduces -td / -tech-detect flag to explicitly opt into technology detection (works with JSONLJSONLA text format where each line is a self-contained, valid JSON object, making it easy to stream, append, and process large datasets line by line without parsing a full JSON array into memory. output).

   2 details in Everything

   └──▷ BREAKING ON UPGRADE

   • !Technology detection is no longer enabled by default; existing workflows that relied on automatic tech detection must now pass -tech-detect explicitly.
10. v1.1.1 Oct 28, 2024 · release notes

    Katana v1.1.1 adds no-clobber output protection, per-host field directory control, htmx endpoint extraction, and auto form-filling for select/textarea.

    └──▷ GET THIS VERSION

    copy

    $ git clone --branch v1.1.1 https://github.com/projectdiscovery/katana.git
    # already have the repo? check out this version:
    $ git checkout v1.1.1

    └──▷ TRY IT
    Prevent katana from overwriting an existing output file when re-running a crawl on the same target.
    copy

    $ katana -u https://example.com -o results.txt -ncb

    Store per-host crawl fields to a dedicated directory for easier post-processing or archiving by host.
    copy

    $ katana -u https://example.com -store-field url -sfd /tmp/katana-fields/

    • Adds -ncb/-no-clobber flag to prevent overwriting existing output files during a crawl.
    • Adds -sfd/-store-field-dir flag to save per-host fields to a custom directory instead of the default location.
    • Adds a response parser that extracts endpoints from htmx attributes, broadening JavaScript-heavy app coverage.
    • Adds automatic form filling for <select> and <textarea> elements to improve crawl depth on forms.
    • Expands JavaScript library fingerprint coverage to improve endpoint parsing across more frameworks.

    +1 more show less

    • Chrome launcher now prioritizes the path specified with -scp over system default Chrome paths in headless mode.

    6 details in Everything

    └──▷ BREAKING ON UPGRADE

    • !Passive crawling flags -ps/-passive and -pss/-passive-source are removed from katana; the capability has been spun out into the separate urlfinder project.
11. v1.1.0 Mar 26, 2024 · release notes

    Adds passive crawling via -passive and passive sources, plus host exclusion filters.

    └──▷ GET THIS VERSION

    copy

    $ git clone --branch v1.1.0 https://github.com/projectdiscovery/katana.git
    # already have the repo? check out this version:
    $ git checkout v1.1.0

    └──▷ USE IT
    Discover endpoints using passive sources only for a target domain.
    shell copy

    katana -u tesla.com -passive -f qurl

    Exclude specific hosts from crawling to reduce noise or respect rate limits.
    shell copy

    katana -u example.com -e 'cdn' -e 'private-ips'

    • -ps, -passive flag enables passive sources for URL discovery.
    • -pss, -passive-source string[] allows selecting specific passive sources (waybackarchive, commoncrawl, alienvault).
    • -e, -exclude string[] flag excludes targets based on filters like cdn, private-ips, cidr, ip, or regex.

    3 details in Everything

12. v1.0.5 Jan 11, 2024 · release notes

    Katana v1.0.5 adds the -disable-redirects flag to control redirect following behavior

    └──▷ GET THIS VERSION

    copy

    $ git clone --branch v1.0.5 https://github.com/projectdiscovery/katana.git
    # already have the repo? check out this version:
    $ git checkout v1.0.5

    └──▷ USE IT
    Scan a target without following redirects, useful for testing raw endpoint responses or avoiding redirect loops.
    shell copy

    katana -url https://example.com -dr

    • Adds the -dr (and --disable-redirects) CLI flag to disable following HTTP redirects

    1 detail in Everything

13. v1.0.4 Sep 14, 2023 · release notes

    Adds -fs custom regex scope, -disable-redirects, and -resume options to Katana v1.0.4

    └──▷ GET THIS VERSION

    copy

    $ git clone --branch v1.0.4 https://github.com/projectdiscovery/katana.git
    # already have the repo? check out this version:
    $ git checkout v1.0.4

    └──▷ TRY IT
    Scope the crawler to specific regex patterns using the new field scope flag.
    copy

    $ katana -fs '(company-staging.io|company.com)' -u company.com

    Resume a paused or interrupted crawl session using the -resume option.
    copy

    $ katana -resume resume.cfg -u https://example.com

    • Added -fs flag for custom regex field scope control
    • Added -dr / -disable-redirects option to stop following redirects
    • Added -resume crawl option to resume scans from a previous run

    3 details in Everything

14. v1.0.3 Aug 1, 2023 · release notes

    Katana v1.0.3 adds DSL response matching/filtering, jsluice parsing, XHR extraction, TLS randomization and existing Chrome instance support

    └──▷ GET THIS VERSION

    copy

    $ git clone --branch v1.0.3 https://github.com/projectdiscovery/katana.git
    # already have the repo? check out this version:
    $ git checkout v1.0.3

    └──▷ USE IT
    Filter crawl URLs where the response status is not 200 or contains specific error strings.
    shell copy

    katana -u https://example.com -fdc "status_code != 200 || body_contains('error')"

    Reuse an active Chrome debugging session to avoid startup overhead.
    shell copy

    katana -u https://example.com -cwu ws://localhost:9222/devtools/browser/1

    • -mdc / -match-condition adds DSL-based advance response matcher to filter URLs by response content.
    • -fdc / -filter-condition adds DSL-based advance response filter to exclude URLs by response content.
    • -tlsi / -tls-impersonate enables experimental client hello (ja3) TLS connection randomization.
    • -jsl / -jsluice enables jsluice based javascript parsing for enhanced endpoint discovery.
    • -fx / -form-extraction extracts form, input, textarea & select elements in jsonl output.

    +2 more show less

    • -xhr / -xhr-extraction extracts XHR request URL and method in jsonl output.
    • -cwu / -chrome-ws-url allows Katana to connect to an existing Chrome browser instance via debugger URL.

    7 details in Everything

15. v1.0.2 Jun 12, 2023 · release notes

    Katana v1.0.2 adds header file input and JSONLJSONLA text format where each line is a self-contained, valid JSON object, making it easy to stream, append, and process large datasets line by line without parsing a full JSON array into memory. output filtering flags

    └──▷ GET THIS VERSION

    copy

    $ git clone --branch v1.0.2 https://github.com/projectdiscovery/katana.git
    # already have the repo? check out this version:
    $ git checkout v1.0.2

    └──▷ TRY IT
    Load headers from headers.txt and omit verbose raw data in JSONL output for cleaner processing.
    copy

    $ katana -H headers.txt -u https://example.com -jsonl -omit-raw -omit-body

    • Adds -H flag to load custom headers from a file for authenticated crawling scenarios
    • Adds -omit-raw / -or flag to exclude raw requests and responses from JSONL output
    • Adds -omit-body / -ob flag to exclude response bodies from JSONL output

    3 details in Everything

16. v1.0.1 Apr 8, 2023 · release notes

    Adds -iqp flag to ignore crawling paths with different query parameters

    └──▷ GET THIS VERSION

    copy

    $ git clone --branch v1.0.1 https://github.com/projectdiscovery/katana.git
    # already have the repo? check out this version:
    $ git checkout v1.0.1

    └──▷ USE IT
    Reduce redundant crawl depth by ignoring query string variations on the same path.
    shell copy

    katana -iqp -u https://example.com

    Crawl a domain without manually specifying http or https in the input.
    shell copy

    katana -u hackerone.com

    • Adds -iqp (-ignore-query-params) flag to prevent crawling the same path when only query parameter values differ
    • Adds support for automatically prepending URL schemes (http/https) to input URLs when not specified

    2 details in Everything

17. v1.0.0 Mar 20, 2023 · release notes

    Katana v1.0 adds request/response correlation, Wappalyzer tech detection, and regex match/filter flags.

    └──▷ GET THIS VERSION

    copy

    $ git clone --branch v1.0.0 https://github.com/projectdiscovery/katana.git
    # already have the repo? check out this version:
    $ git checkout v1.0.0

    └──▷ USE IT
    Use specific DNS resolvers to bypass default resolution or test internal infrastructure.
    shell copy

    katana -u https://example.com -r ./resolvers.txt

    Filter crawl results to only include URLs matching a specific pattern (e.g., admin panels).
    shell copy

    katana -u https://example.com -mr 'admin.*'

    • -r / -resolvers flag allows specifying a list of custom DNS resolvers (file or comma-separated)
    • -mr / -match-regex and -fr / -filter-regex flags enable regex match and filter on discovered endpoint URLs
    • JSON output now includes request and response fields containing the raw HTTP request, response headers, status code, body, and detected technologies via Wappalyzer

    3 details in Everything

    └──▷ BREAKING ON UPGRADE

    • !The JSON output structure has changed: flat fields like endpoint, source, tag, and attribute are now nested under request, and a new response field containing status_code, headers, body, and technologies is added.
18. v0.0.3 Jan 13, 2023 · release notes

    Katana v0.0.3 adds -scp for local chrome binary, -elog for error logs, -hc health-check, and raw HTTP store support

    └──▷ GET THIS VERSION

    copy

    $ git clone --branch v0.0.3 https://github.com/projectdiscovery/katana.git
    # already have the repo? check out this version:
    $ git checkout v0.0.3

    └──▷ TRY IT
    Scan a target using a locally installed Chrome binary instead of the system default
    copy

    $ katana -u https://target.com -scp /usr/local/bin/chrome

    Redirect error logs to a file for debugging headless browsing issues
    copy

    $ katana -u https://target.com -elog katana-errors.log

    • Added the --chrome-path flag (-scp) to specify a local Chrome binary for headless browsing
    • Added the --error-log flag (-elog) to write error logs to a file instead of stderr
    • Added the --health-check flag (-hc) for self-diagnostic functionality

    3 details in Everything

19. v0.0.2 Nov 11, 2022 · release notes

    Adds -no-sandbox and -headless-options flags to control headless Chrome execution

    └──▷ GET THIS VERSION

    copy

    $ git clone --branch v0.0.2 https://github.com/projectdiscovery/katana.git
    # already have the repo? check out this version:
    $ git checkout v0.0.2

    └──▷ TRY IT
    Run Katana on a localhost target where Chrome's default sandbox cannot be initialized (e.g. CI/CD with root)
    copy

    $ katana -u https://target.com --no-sandbox

    Inject custom browser capabilities (e.g. disable GPUGPUA graphics processing unit, a processor that performs many calculations in parallel. AI workloads use GPUs to speed up model training and inference. acceleration or set window size) during a crawl
    copy

    $ katana -u https://target.com --headless-options="--disable-gpu --window-size=1920,1080"

    • Adds --no-sandbox flag to run the headless browser in root environments where sandboxingsandboxingRunning code inside a restricted environment that limits what it can access or change. The isolation reduces the impact of mistakes or untrusted code, depending on the sandbox's controls. is unavailable
    • Adds --headless-options argument to pass arbitrary Chrome command-line flags directly to the underlying headless instance

    2 details in Everything