agentleFS
Sn4kyGitSn4kyGit/qa-agent-blueprint/.claude/skills/recon/SKILL.md

recon

Skill2 starsChanged 7 days agoApache-2.0View on GitHub

Run pipeline phase 2 (recon) for a ticket or area — probe script (Playwright or WebdriverIO) that observes pages and writes recon.json with the real selectors, console errors and screenshots into the run dir. Use for "/recon <KEY-123>", "erkunde die Seite vor dem Test", or whenever a spec would otherwise be written blind. Not for live exploratory testing — /exploratory-testing.

32 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use reporting, osint and 3 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • GitHubSkill

    reporting

    Report formatting guidelines for gh-aw workflows (header levels, collapsible details, run-ID links)

    Also in 3 projects that use this

  • JosBen74Skill

    osint

    OSINT-sökning och informationsinhämtning från öppna källor för Länsstyrelsen VG

    Also in 4 projects that use this

  • samklee1Skill

    web-pentest

    Orchestrates authorized web app pentesting on Kali Linux using Opus sub-agents with difficulty-aware planning (TDA/EGATS), browser-first exploration, iterative verification, two-layer review (per-sub-agent + phase-level), optional cross-model gap audit (GPT/codex via codex-plugin-cc), and enforced KB updates. Trigger on: pentesting, penetration testing, web app security assessment, vulnerability assessment, OWASP testing, PTES, attack surface mapping, API security testing, pentest reporting, or 'help me test this app'. NOT for network-only pentesting, malware analysis, reverse engineering, or CTFs.

    Also in 3 projects that use this

  • SentrySkill

    create-pr

    Create pull requests following Sentry conventions. Use when opening PRs, writing PR descriptions, or preparing changes for review. Follows Sentry's code review guidelines.

    Also in 3 projects that use this

  • NVIDIASkill

    research

    Before designing a feature, research how comparable libraries solve the problem, check best practices, known pitfalls, and upstream changes. Use when starting a new feature, investigating an unfamiliar area, or the user asks to research something.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

1 rule. 0 are common, 0 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 27 edited versions we read keep these.

  • 0

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Expiry check: if a previous recon.json for this area exists and the new run shows different controls, say so — specs built from the old …

    line 42

“Common” is measured: kept by at least 6 in 10 of the 27 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

/recon — observe before you assert

A spec written without seeing the page guesses its selectors — the second-most common cause of worthless CLI results (docs/pipeline/00-overview.md). This skill produces recon.json: real controls with role and accessible name, form fields, console errors, failed requests, screenshots — for a fraction of the tokens an MCP session costs. Full phase contract: docs/pipeline/02-recon.md.

Steps

  1. Resolve scope. Ticket key (or --project <KEY> --product <PRODUCT> headless). Read the effective project file: "Execution mode" (on cli-first, reuse an existing recon.json unless expired — do not re-probe by habit) and "Test assets" → Framework (playwright → recon.spec.ts; webdriverio → recon.wdio.spec.js).
  2. Run dir. Reuse the active run dir (evidence/<KEY>/<TICKET>/<YYYYMMDD-HHMMSS>/) when called mid-pipeline; standalone, create a fresh one. Never overwrite an existing recon.json in a reused dir — note the refresh in progress.md instead.
  3. Paths. From intake.json → affected_paths when present; else from the ticket plus confirmed Areas in flow-map.md. Never derive paths by scanning the app source or treat Unconfirmed candidates as established routes.
  4. Ensure the probe exists. Templates ship with scaffold-suite (templates/probes/). Suite has no probes/ copy yet → copy the matching variant there (it is infrastructure, commit it; part files and recon output stay in the run dir).
  5. Run per the probe README: RUN_DIR, BASE_URL from the project's stage URL, credentials injected via qa-secrets.mjs run — never literal, never wholesale dotenv. Logged-in recon via STORAGE_STATE when the flow needs it.
  6. Merge: node probes/merge-recon.mjs <RUN_DIR>. Only its stdout summary enters the context — never the part files or the full recon.json (token rule, docs/pipeline/02-recon.md).
  7. Report back, 3–5 lines: pages probed, control/field counts per page, observations (console errors, ≥400 responses, a11y, slow pages), path of recon.json. On a page unreachable → that is a finding, not a retried run.
  8. Expiry check: if a previous recon.json for this area exists and the new run shows different controls, say so — specs built from the old recon are outdated (re-run recon, don't patch specs onto a changed page).

This skill writes no summary of its own: when the calling flow writes the run's execution_summary.json, recon belongs in its skills_used if this skill actually ran (.claude/rules/core/evidence-policy.md).

MCP instead of probe

Unclear interaction, non-reproducible bug, visual assessment, or Mode: mcp-only → skip the probe and charter test-executor (exploratory-testing) instead. The session still produces the phase artifact: at its end the executor writes recon.json into the run dir with what it observed — pages, controls (role + accessible name), form fields, console errors, failed requests — in the probe output format (docs/pipeline/02-recon.md, "source": "mcp"). Both recon paths are producers of recon.json; the expiry check (step 8) applies to MCP output too.

What this skill never does

Load recon.json or part files wholesale into the context. Assert behavior (the probe claims reachability only). Write credentials anywhere. Scan the application source for paths.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.