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 skills/evilfreelancer/secs/recognizing-deceptionnpx skills add EvilFreelancer/secs --skill recognizing-deceptiongit clone --depth 1 https://github.com/EvilFreelancer/secsWhat 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 | $0.00119 | $0.01494 |
| Opus 5 | $0.00060 | $0.00747 |
| Sonnet 5 | $0.00024 | $0.00299 |
| Haiku 4.5 | $0.00012 | $0.00149 |
Grade C, and why
recognizing-deception scanned grade C 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 3d 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.
Reaches for credential fileshighPrivilege escalation
SSH keys, cloud credentials, git-credentials, .npmrc, /etc/shadow: reading these is how a config file becomes a credential leak.
| Canary credential (AWS/Slack/kube/Azure cert) | Valid **format** but tied to no real identity; sitting in `~/.aws/credentials`, a repo `.env`, `/root/.kube/config`; never appears in prior legitimate API/auth logs | How it starts
The opening of the file, as written. The whole thing — 100 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Recognizing Deception
Modern defenders plant tripwires whose only purpose is to fire when an intruder touches them: a credential that has no legitimate use, an account that never logs on, a document that phones home when opened. Because they have zero legitimate use, any interaction is a definitive, high-fidelity compromise alert — which makes them the cheapest way for a blue team to catch a red team. Recognizing them is an OPSEC skill: the goal is not to disarm the defense but to avoid stepping on it and burning the engagement. The core instinct is provenance. Access you cannot attribute to a specific step you took is a trap until proven otherwise.
This skill is advisory judgement layered over other work — it does not add target
interaction beyond what you are already authorized to do. It pairs tightly with
credential provenance in maintaining-engagement-state: log where every
credential and access came from, so "I don't know where this came from" becomes a
visible stop signal rather than a shrug.
When to Use
- Before acting on found credentials, keys, or tokens of uncertain origin
- Before Kerberoasting or AS-REP roasting an account that looks conveniently exposed
- Before opening an enticingly-named file or browsing a suspiciously discoverable share
- Before using an AWS/Slack/kube credential discovered in a repo or on a host
- Any time access appears without a step you can point to that produced it
When NOT to Use
- Deploying deception as the defender — that is the inverse discipline and is out of scope here
- Recording where a credential came from — use
maintaining-engagement-state; this skill decides whether to trust it, that one stores the provenance - The active AD attack itself — use
attacking-active-directory; consult this skill before you pull the trigger - Password attacks on a hash you already hold — use
cracking-passwords; check the account is not bait first - Cloud credential abuse — use
exploiting-cloud-platforms; validate the key is not a canary first
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.
- 3d ago First seen · 100 lines · 119 tokens per session scan C 8b04a8acdc29
recognizing-deception is a skill published in the GitHub repository EvilFreelancer/secs (10 stars, last pushed 24d ago), licensed Apache-2.0. It adds 119 tokens to every session and 1,494 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it C with 1 finding (reaches for credential files). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
Blue Team Defense & Hardening
System hardening, detection engineering, security baseline monitoring, patch management, defense-in-depth architecture, and security posture improvement.
analyzing-email-headers-for-phishing-investigation
Parse and analyze email headers to trace the origin of phishing emails, verify sender authenticity, and identify spoofing through SPF, DKIM, and DMARC validation.
analyzing-golang-malware-with-ghidra
Reverse engineer Go-compiled malware using Ghidra with specialized scripts for function recovery, string extraction, and type reconstruction in stripped Go binaries.
analyzing-prefetch-files-for-execution-history
Parse Windows Prefetch files to determine program execution history including run counts, timestamps, and referenced files for forensic investigation.
analyzing-ios-app-security-with-objection
Performs runtime mobile security exploration of iOS applications using Objection, a Frida-powered toolkit that enables security testers to interact with app internals without jailbreaking. Use when assessing iOS app security posture, bypassing client-side protections, dumping keychain items, inspecting filesystem…
add-tool
Use when adding a new cybersecurity tool to this installer. Walks through editing the right module file, adding to toolsconfig.json, running validators, and syncing MCP server data if needed. Triggers on phrases like "add tool", "add ", "register a new tool", "include X in the installer".