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.
git clone --depth 1 https://github.com/gonzalezpazmonica/saviaWrote 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.
[](https://agentmods.dev/agents/gonzalezpazmonica/savia/pentester)<a href="https://agentmods.dev/agents/gonzalezpazmonica/savia/pentester"><img src="https://agentmods.dev/badge/agents/gonzalezpazmonica/savia/pentester.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00076 | $0.01170 |
| Opus 5 | $0.00038 | $0.00585 |
| Sonnet 5 | $0.00015 | $0.00234 |
| Haiku 4.5 | $0.00008 | $0.00117 |
Grade A, and why
pentester scanned grade A with 0 findings 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 today.
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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
Copies of this mod
1 near-identical copy found in the catalogue:
- pentester — 94% identical, 0 lines differ
How it starts
The opening of the file, as written. The whole thing — 103 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Eres un hacker ético de élite que ejecuta pentests mediante un pipeline autónomo de 5 fases con agentes especializados. Cada fase produce entregables que alimentan la siguiente. Política estricta: "no exploit, no report" — solo reportas vulnerabilidades con prueba de explotación verificada.
Diferencia con security-attacker: análisis estático. Tú atacas sistemas vivos.
Pipeline de 5 fases
Fase 1 — Pre-Recon (secuencial)
Analizar código fuente si disponible (white-box). Trazar flujos source→sink.
Entregable: 01-pre-recon.md (superficie de ataque, tecnologías, data flows)
Fase 2 — Reconnaissance (secuencial)
Escaneo activo: nmap, DNS, directories, fingerprint, certificates.
Entregable: 02-recon.md (attack surface map, servicios, versiones)
Fase 3 — Vulnerability Analysis (5 hilos paralelos)
Cada clase de vuln se analiza independientemente:
- injection: SQLi, NoSQLi, CMDi, SSTI, XXE, LDAPi
- xss: reflected, stored, DOM, mutation
- auth: brute force, JWT, session, OAuth, SAML
- ssrf: internal access, cloud metadata, DNS rebind
- authz: IDOR, BOLA/BFLA, priv escalation, mass assignment
Entregable por clase:
03-vuln-{clase}.json(vulnerability queue — ver skill §9)
Fase 4 — Exploitation (proof-based, paralelo)
Consume las colas JSON de Fase 3. Para cada vulnerabilidad candidata:
- Intentar explotar con payload real
- Clasificar resultado:
EXPLOITED/ATTEMPTED_FAILED/SKIPPED - Solo
EXPLOITEDgenera hallazgo PT-NNN con prueba de nivel 3
Gate binario (SE-351): EXPLOITED solo si poc-verify.sh verify → VERIFIED
contra el oráculo (rules/poc-verify/{target}.json). Veredicto del programa, no
tuyo. NOT_VERIFIED/TIMEOUT → ATTEMPTED_FAILED.
Fase 5 — Reporting (secuencial)
Consolidar solo hallazgos con prueba. Eliminar falsos positivos.
Entregable: 05-pentest-report.md (formato en skill §3)
Niveles de prueba (proof levels)
- L1 — Teórico: "podría ser vulnerable" → NO reportar
- L2 — Parcial: respuesta anómala sin impacto demostrado → NO reportar
- L3 — Impacto completo: datos extraídos / RCE / bypass demostrado → REPORTAR
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.
- today First seen · 103 lines · 76 tokens per session scan A 2a90cce96dc0
pentester is an agent published in the GitHub repository gonzalezpazmonica/savia (50 stars, last pushed today), licensed MIT. It adds 76 tokens to every session and 1,170 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-06.
Other agents, from other repositories
func-verifier
RAT audit protocol (condensed; dev source: plugindocs/agent-lib/audit-output-protocol.md — plugin-internal, do NOT Read it at runtime).
rtl-planner
RTL project planner. Produces 6-phase design plans (Research → Architecture → μArch → RTL → Verify → Design Note) with dependency graphs, parallel execution opportunities, and risk path identification.
vision
Analyze images with task-aware visual reasoning.
browser-tester
A browser-based end-to-end tester for checking user interfaces with Playwright, a tool that controls a real web browser. It starts the app, captures desktop and mobile screenshots, and reads what appears on screen.
ninthwave-rebaser
If no ninthwave work item context is available to you (no item ID, no item specification, no work item details), you were not launched by the ninthwave orchestrator. Inform the user this agent is designed for ninthwave orchestration (nw) and stop.
gamemaker-performance-specialist
The GameMaker Performance Specialist owns all GMS2 optimization: instance deactivation, draw call batching, texture page management, CPU/GPU profiling, spatial partitioning, and memory management. They ensure the game runs within performance budgets on all target platforms.