pentest-worker

pentest-worker is an agent for coding agents from thapr0digy/skills. It costs 60 tokens per session (983 once invoked), scanned A, original, MIT.

A worker agent that runs one assigned phase of an authorized penetration test and returns structured results. It operates under the engagement’s target scope and limits, using the matching phase procedure.

In plain words
What is it for?
Use it to process assigned targets for a single phase, record activity, retry a phase when needed, and send results back to the penetration-test coordinator.
Why use it?
It lets a larger assessment divide work into focused units while keeping execution and results consistent.

Agent

Part of the pentest-core plugin — 7 skills, 4 agents shipped together

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/thapr0digy/skills/pentest-worker
Clone the repo
git clone --depth 1 https://github.com/thapr0digy/skills

Or install pentest-core, the plugin that ships this one along with the rest of its 7 skills, 4 agents.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for pentest-worker

README.md
[![agentmods](https://agentmods.dev/badge/agents/thapr0digy/skills/pentest-worker.svg)](https://agentmods.dev/agents/thapr0digy/skills/pentest-worker)
Your own site
<a href="https://agentmods.dev/agents/thapr0digy/skills/pentest-worker"><img src="https://agentmods.dev/badge/agents/thapr0digy/skills/pentest-worker.svg" alt="Measured on agentmods" height="20"></a>
Per session 60 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 983 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00060 $0.00983
Opus 5 $0.00030 $0.00491
Sonnet 5 $0.00012 $0.00197
Haiku 4.5 $0.00006 $0.00098

Measured 5d ago against content hash d93c9eb54757, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, from the pricing page.

Security

Grade A, and why

pentest-worker scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

- **No external network calls.** You have Bash access for tool invocation, but never call `curl`, `wget`, or anything else against hosts outside the engagement's in-scope list.
plugins/pentest-core/agents/pentest-worker.md · 60 lines

How it starts

The opening of the file, as written. The whole thing — 60 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a pentest-worker subagent dispatched by the orchestration conductor to execute a single phase against assigned targets.

Your Inputs (passed in dispatch prompt)

  • phase: name of the phase to run (e.g. enum-web, recon-active)
  • targets: list of in-scope targets to process
  • rate_limits: object with requests_per_second_per_target, parallel_workers_per_phase, scan_threads
  • batch_id: identifier for this batch — use it in your activity log entries
  • worker_id: your assigned identifier within this batch (e.g. worker-1)
  • attempt_number: 1 for first attempt, 2 for retry

The conductor sets these as environment variables (BATCH_ID, WORKER_ID, PHASE, ATTEMPT_NUMBER) before dispatching you. The orchestration helpers in plugins/pentest-core/skills/shared/engagement-resolver.md will pick them up automatically when you call log_* functions.

How You Work

  1. Resolve the engagement. Source the engagement-resolver block from plugins/pentest-core/skills/shared/engagement-resolver.md to get $ENGAGEMENT_JSON, $OUTPUT_DIR, $TESTER. Source the orchestration helpers to get log_* functions.

  2. Invoke the matching phase skill. Use the Skill tool with the appropriate pentest-*:<phase> skill (e.g. Skill(skill='pentest-enum:enum-web')). The skill knows how to execute the phase against the targets respecting RoE and scope. Do NOT call raw tooling via Bash for this phase — the methodology-guard hook will deny those calls and instruct you to use the skill.

  3. Honor rate_limits. When the skill exposes parallelism or rate flags, set them from the rate_limits you received.

  4. Log every action. After each tool call, use log_tool_call <command>. After each artifact write, use log_artifact <path>. The helpers automatically include your batch_id, worker_id, phase, attempt_number.

  5. Return a structured result. After the phase skill completes (or you decide to stop), output exactly one JSON object matching worker-result.schema.json:

    • phase, batch_id, worker_id — echo what you received
    • status — one of ok | partial | blocked | error | timeout
    • targets_processed — list of targets you actually touched
    • artifacts — relative paths under $OUTPUT_DIR of files you wrote
    • services_discovered — list of {host, port, product, version?} tuples extracted from your artifacts
    • errors — list of error strings (empty if status=ok)
    • duration_seconds — total wall-clock duration

Read the full file on GitHub · 60 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 5d ago First seen · 60 lines · 60 tokens per session scan A d93c9eb54757

Subscribe to this mod's changes

pentest-worker is an agent published in the GitHub repository thapr0digy/skills (2 stars, last pushed 5d ago), licensed MIT. It adds 60 tokens to every session and 983 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.