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.
npx agentmods add commands/jonase47/ccpr/p6-pentest-recongit clone --depth 1 https://github.com/jonase47/ccprWrote 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/commands/jonase47/ccpr/p6-pentest-recon)<a href="https://agentmods.dev/commands/jonase47/ccpr/p6-pentest-recon"><img src="https://agentmods.dev/badge/commands/jonase47/ccpr/p6-pentest-recon.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 | $0.00000 | $0.00837 |
| Opus 5 | $0.00000 | $0.00418 |
| Sonnet 5 | $0.00000 | $0.00167 |
| Haiku 4.5 | $0.00000 | $0.00084 |
Grade A, and why
p6-pentest-recon 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 2d 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.
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.
How it starts
The opening of the file, as written. The whole thing — 89 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/p6-pentest-recon – Reconnaissance
Collects publicly available information: tech fingerprinting, exposed endpoints, debug remnants.
Argument: $ARGUMENTS = [target URL/system]
If provided: Focus recon on the specified target. If not provided: Ask about the pentest scope.
Prerequisites
- System deployed or running locally
- ARCHITECTURE.md present
Agent
- Type: pentester
- Model: sonnet
Context (Orchestrator prepares)
Orchestrator reads in advance and delivers inline:
- From ARCHITECTURE.md: System components, technologies
- From SECURITY.md: Known threats from the threat model
- From SECURITY_AUDIT_REPORT.md: Audit findings as starting points (if present)
Prompt Template
Goal: Reconnaissance for: [target/system]
Known Architecture: [inline from ARCHITECTURE.md]
Output Format:
# Information Type Found Risk Note Categories: Tech Stack, Exposed Endpoints, Admin Areas, Debug Info, Error Leaks Max. 15 findings
Constraints:
- Reconnaissance ONLY – no active attacks
- Only against own system
- Gather information, do not exploit
Orchestrator Checkpoint
- Recon covers all attack surfaces?
- Found information usable as basis for further pentest phases?
Write Detail File
Write the result to docs/quality/pentest_recon.md (overwrite). Frontmatter:
---
phase: P6
subskill: pentest-recon
status: active
last_updated: <DD.MM.YYYY>
---
Body sections: ## Scope, ## Findings (recon results table), ## Notable Disclosures.
Update Sub-Index
Update docs/quality/PENTEST.md (the pentest sub-index, created by /p6-pentest if missing):
- Set
**Last Updated:** <DD.MM.YYYY>. - In its Detail Files table: ensure a row for
[pentest_recon.md](pentest_recon.md)with statuscomplete. - Lift any sensitive disclosure (exposed admin endpoints, leaked metadata, etc.) into Open Risks of the sub-index.
- Do not edit
QA.mddirectly.
Handover Epilogue
Before writing. docs/HANDOVER.md is capped — the file states its own limit in its header
(default: ≤5 KB / ~150 lines). Two rules follow from that, and neither is optional:
- Replace this command's previous epilogue block, do not append a second one. Stacking is what pushes the file over; one skill run has been measured adding 1021 B, ~20 % of the cap.
- If the file is already near its cap, shorten before you add. Reading the cap sentence is not
the same as measuring: check the actual size, and when there is no room, condense existing content
or hand the user
/cleanupinstead of growing the file further.
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.
- 2d ago First seen · 89 lines · 0 tokens per session scan A e53ecbaba944
p6-pentest-recon is a command published in the GitHub repository jonase47/ccpr (1 stars, last pushed today), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 837 tokens. 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 2026-09-03.
Other commands, from other repositories
sdd-implement
Execute a task list in dependency order, tests before implementation for each task, halting on any non-parallel failure instead of pressing ahead.
wiki-tune
Tune wiki schema and conventions interactively.
experiment
Design A/B test with statistical rigor. Usage: /mindforge:experiment [feature] [--metric conversion|engagement|revenue] [--duration 2w].
pseo-init
Use when: kullanıcı "yeni proje", "init", "proje kur", "yeni domain ekle", "scaffold" gibi ifadeler kullanır ya da /pseo-init çağırırsa. Also use when: portföye yeni bir SEO projesi alındı, projects/{slug}/ klasörü ve project.config.json dosyasının schemaya uygun ilk hâli üretilecek; brief'te slug + domain + market…
forge-insights
Analyze past sessions for error patterns, file activity, and recommendations.
forge-context-status
Report on current context window usage, cache health, and compaction recommendation. Read-only — does not compact.