Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsBlogPro
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Authors
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges
  • Chrome Extension
  • Skill Manager

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

ProTermsPrivacyRefunds
Back to skills

Site Fleet

ASecurity

This skill should be used when the user asks to analyse/recon MULTIPLE sites at once — "recon these sites", "map the API surface of A, B and C", "run beacon over this list of URLs", or runs /beacon:fleet. Recons sources one at a time through site-analyst, tracks them in a durable ledger, and gates completion. For a single site, use site-recon instead.

2 stars
0 votes
0 copies
0 views
Added 9/19/2026
ai-agentspythongobashapi

Works with

terminalapi

Security Analysis

A100/100

Scanned 9/19/2026

$npx -y skills add neotherapper/claude-plugins --skill site-fleet --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Site Fleet?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Site Fleet
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/neotherapper-site-fleet/badge)](https://www.skillsdirectory.com/skills/neotherapper-site-fleet)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
Files
SKILL.md
---
name: site-fleet
description: This skill should be used when the user asks to analyse/recon MULTIPLE sites at once — "recon these sites", "map the API surface of A, B and C", "run beacon over this list of URLs", or runs /beacon:fleet. Recons sources one at a time through site-analyst, tracks them in a durable ledger, and gates completion. For a single site, use site-recon instead.
license: MIT
metadata:
  version: "0.9.0"
  author: Georgios Pilitsoglou
---

# site-fleet — Sequential fleet orchestration

Recon a LIST of sources one at a time, reusing site-recon per source, with a durable
ledger that survives compaction and a deterministic completeness gate.

**Reuses, never replaces:** each source is a full site-recon (Subsystem A). This skill only
adds the ledger, the sequential loop, and the completeness sweep. One source at a time — no
parallelism (that is the deferred B2 subsystem).

## Entry — branch on arguments (REQUIRED)

- **With URLs / a file arg** → start a new fleet:
  ```bash
  python3 "${CLAUDE_PLUGIN_ROOT}/skills/site-recon/scripts/fleet.py" init {url1} {url2} …
  ```
  (A file of URLs, one per line: expand it to the arg list first.) Record the printed
  `[FLEET:{ledger}]`. If it prints `[FLEET-ERROR] … already active`, do NOT retry init —
  go to resume.
- **With no args, OR after `[FLEET-ERROR] already active`** → resume; never call `init`:
  ```bash
  python3 "${CLAUDE_PLUGIN_ROOT}/skills/site-recon/scripts/fleet.py" pending
  ```

## The loop (sequential)

Repeat until `fleet.py pending` prints nothing:

1. Take the next slug from `fleet.py pending` (it also re-arms a paused fleet).
2. **Reconcile:** if that source's `docs/sites/{slug}/research/INDEX.md` is already complete
   (`python3 …/okf_validate.py --is-complete <INDEX>`), mark it and skip:
   `fleet.py update {slug} --status complete --verdict complete`.
3. Otherwise mark it in flight: `fleet.py update {slug} --status reconning`.
4. **Recon the whole source (Phases 1–12) in ONE context.** Dispatch the `site-analyst`
   subagent for `{url}` and await its completion — a foreground `site-analyst` agent can run
   the entire recon, including the browser phases, end to end. This is the primary path.
   (Fallback only: if a future environment denies subagents Bash/browser access, run the
   `site-recon` skill for `{url}` yourself in the main session instead.) Do NOT split
   Phases 1–9 from 10–11 across contexts — that re-opens the content seam (prohibited in B1).
5. Record the outcome:
   - completed (INDEX flipped `status: complete`) → `fleet.py update {slug} --status complete --verdict complete`
   - blocked for a concrete reason (e.g. auth wall) → `fleet.py update {slug} --status blocked --verdict blocked:{reason}`
   - errored / produced nothing → `fleet.py update {slug} --status inconclusive --verdict inconclusive`, then **retry once**; if still not complete, `waive` it so it reaches a terminal status (`inconclusive` is NOT terminal — leaving it inconclusive would make `fleet.py pending` re-list it forever): `fleet.py waive {slug} --reason inconclusive-after-retry`. Each `--status inconclusive` update bumps the row's durable `retries` count in the ledger (survives compaction) — the retry policy is bounded to **one** retry, so once `retries` reaches 2 for a slug, waive it rather than retrying again; do not rely on in-context memory of how many times you've retried, check `retries` in the ledger.

## Close

- `python3 …/fleet.py sweep` → prints `[FLEET-COMPLETE]` or `[FLEET-INCOMPLETE:{slug}:{reason}]`.
- Report a slug → status table and the incomplete list.
- **Pausing across sessions:** before an intentional mid-fleet stop, run `fleet.py pause`
  (preserves the resume handle; the Stop gate will not block). Resume next session with
  `/beacon:fleet` (no args) or `fleet.py pending`.
- **A source that genuinely cannot complete:** `fleet.py waive {slug} --reason {why}`.
- **Abandon the fleet:** `fleet.py close`.

If you stop with unresolved sources and have not paused, the `fleet-sweep.sh` Stop hook blocks
with `[FLEET-SWEEP-PENDING:…]` — finish, waive, pause, or close.

Attribution

neotherapperneotherapper
View sourceSee grades on GitHubMore from neotherapper →
SSkills Directory ProSkills Directory

Get any skill into Claude in one click.

Download any skill as a ZIP for Claude.ai, Claude Desktop, or .claude/skills. $9/mo.

See Pro

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills Directory ProSkills Directory

Get any skill into Claude in one click.

Download any skill as a ZIP for Claude.ai, Claude Desktop, or .claude/skills. $9/mo.

See Pro

Related Skills

Caveman

Ultra-compressed communication mode that cuts output tokens while keeping technical accuracy. Levels: lite, full, ultra and the wenyan variants. Use for /caveman, "caveman mode", "talk like caveman", "be brief" or "less tokens".

1087401 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

697551 votes

Writing Skills

Create and manage Claude Code skills in HASH repository following Anthropic best practices. Use when creating new skills, modifying skill-rules.json, understanding trigger patterns, working with hooks, debugging skill activation, or implementing progressive disclosure. Covers skill structure, YAML frontmatter, trigger types (keywords, intent patterns), UserPromptSubmit hook, and the 500-line rule. Includes validation and debugging with SKILL_DEBUG. Examples include rust-error-stack, cargo-dep...

3931 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3421 votes

catchup

Recovers the conversation and failed tool calls of a previous Codex, Amp, Claude Code, Antigravity, Cline, Copilot CLI, Cursor, DeepSeek Harness, Grok Build, Kimi, OpenCode, Pi Agent, or ZCode session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", asks to recover/summarize a previous session before continuing, or asks to diagnose or report a catchup failure. Do NOT use for the current conversation, git history, or any non-agent log.

741 votes
View all in ai-agents →