Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/kalpmodi/akiranpx agentmods add skills/kalpmodi/akira/403-bypassWrote 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/kalpmodi/akira/403-bypass)<a href="https://agentmods.dev/skills/kalpmodi/akira/403-bypass"><img src="https://agentmods.dev/badge/skills/kalpmodi/akira/403-bypass/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/kalpmodi/akira/403-bypass"><img src="https://agentmods.dev/badge/skills/kalpmodi/akira/403-bypass.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00090 | $0.15881 |
| Opus 5 | $0.00045 | $0.07941 |
| Sonnet 5 | $0.00018 | $0.03176 |
| Haiku 4.5 | $0.00009 | $0.01588 |
Grade F, and why
403-bypass scanned grade F with 4 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Sends data to an external URLmediumData exfiltration
A POST to an outside endpoint may be telemetry or may be exfiltration; either way the mod talks to somewhere, and you should know where.
CODE=$(curl -sk -X POST "$TARGET" -H "Content-Type: application/json" -d "$PAYLOAD" -o /dev/null -w "%{http_code}") Downloads and executes remote codehighSupply chain
curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.
curl -sk "https://web.archive.org/cdx/search/cdx?url=target.com/restricted&output=json&limit=5" | python3 -m json.tool Cloud metadata endpointhighServer-side request forgery
One request to 169.254.169.254 can return temporary IAM credentials.
"http://169.254.169.254/latest/meta-data/" # AWS metadata Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -ski "https://target.com/restricted" -o /dev/null -w "HTTP: %{http_code} | Size: %{size_download} | Time: %{time_total}s\n" How it starts
The opening of the file, as written. The whole thing — 1,309 lines — stays where its author put it; the contents beside it link to each section on GitHub.
403 Bypass
Overview
A 403 doesn't always mean truly inaccessible. Misconfigured reverse proxies, CDNs, WAFs, and backend servers enforce access controls at different layers - creating parser-level gaps. This skill covers every known bypass class, from basic header tricks to conference-grade research techniques (DEF CON 2024, Black Hat 2024/2025, Orange Tsai's parser logic work, PortSwigger cache research).
Goal: Confirm whether a 403 is a hard block or a bypass-able misconfiguration, then document evidence for reporting.
Phase 0: Smart Intake
source ~/.claude/skills/_shared/phase0.sh
source ~/.claude/skills/_shared/signals.sh
p0_init_vars "$1"
p0_state_gate "HARVEST" || exit 0
p0_read_relay recon
p0_read_memory
p0_read_hypotheses
TECH_STACK=$KNOWN_TECH # alias for downstream references
echo "=== PHASE 0 403-BYPASS INTAKE: $TARGET ==="
echo "State: $STATE | WAF: $WAF"
echo "Tech stack: $TECH_STACK"
echo "Target endpoints for bypass: $(echo "$INTERESTING_ENDPOINTS" | wc -l)"
echo "Top hypothesis: $TOP_HYPO_LABEL [$TOP_HYPO_PROB%]"
echo "ATW flagged (avoid): ${ATW_FLAGGED:-none}"
Execution Manifest
Build targeted manifest: one item per bypass technique class per target endpoint. Priority is determined by WAF type detected.
# WAF-to-priority mapping:
# Cloudflare -> c02 (header), c05 (double-encode), c12 (automated), c17b (origin IP direct hit)
# Akamai -> c10 (smuggling OPTIONS+line-folding CVE-2025-32094), c11 (cache deception), c02
# AWS WAF -> c17 (ALB direct), c02 (header), c04 (path manipulation)
# Nginx+Apache-> c07 (path confusion CVE-2025-0108), c08 (off-by-slash), c15 (mod_proxy CVE-2024-38473)
# IIS -> c18 (tilde enum + ADS), c22b (Apache ? bypass if applicable)
# unknown -> c02, c04, c09, c12 (automated sweep)
MANIFEST=$(cat << 'MANIFEST_EOF'
{
"phase": "403-bypass",
"generated_at": "YYYY-MM-DD HH:MM",
"items": [
{"id":"b01","tool":"baseline","target":"<403-endpoint>","reason":"fingerprint WAF/CDN/server before trying anything","priority":"MUST","status":"pending","skip_reason":null},
{"id":"b02","tool":"header-bypass","target":"<403-endpoint>","reason":"X-Forwarded-For/X-Original-URL IP spoofing","priority":"MUST","status":"pending","skip_reason":null},
{"id":"b03","tool":"path-manipulation","target":"<403-endpoint>","reason":"encoding/case/extension tricks bypass string-match ACL","priority":"MUST","status":"pending","skip_reason":null},
{"id":"b04","tool":"hop-by-hop-strip","target":"<403-endpoint>","reason":"strip auth headers via Connection: header","priority":"SHOULD","status":"pending","skip_reason":null},
{"id":"b05","tool":"double-url-encoding","target":"<403-endpoint>","reason":"WAF decodes once, backend decodes twice (DEF CON 2024)","priority":"SHOULD","status":"pending","skip_reason":null},
{"id":"b06","tool":"method-fuzzing","target":"<403-endpoint>","reason":"ACL may only block GET, not PUT/POST/TRACE","priority":"SHOULD","status":"pending","skip_reason":null},
{"id":"b07","tool":"nomore403","target":"<403-endpoint>","reason":"330+ automated bypass techniques","priority":"MUST","status":"pending","skip_reason":null},
{"id":"b08","tool":"waf-specific-bypass","target":"<403-endpoint>","reason":"targeted technique for detected WAF type","priority":"MUST","status":"pending","skip_reason":"skip if WAF=unknown"},
{"id":"b09","tool":"h2c-smuggling","target":"<403-endpoint>","reason":"h2c upgrade causes proxy to stop inspecting requests","priority":"IF_TIME","status":"pending","skip_reason":null},
{"id":"b10","tool":"prototype-pollution","target":"<403-endpoint>","reason":"Node.js/Express auth bypass via __proto__","priority":"IF_TIME","status":"pending","skip_reason":"skip if no Node.js in tech stack"}
]
}
MANIFEST_EOF
)
jq --argjson m "$MANIFEST" '.scalpel.active_manifest = $m' $SESSION > /tmp/s.json && mv /tmp/s.json $SESSION
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.
- 9d ago First seen · 1,309 lines · 90 tokens per session scan F c3678a0ccdbf
403-bypass is a skill published in the GitHub repository kalpmodi/akira (21 stars, last pushed 1mo ago), licensed MIT. It adds 90 tokens to every session and 15,881 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it F with 4 findings (sends data to an external url, downloads and executes remote code, cloud metadata endpoint). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
hunt-llm-ai
Hunt LLM/AI feature bugs — prompt injection, indirect injection, exfiltration viatool-use/markdown, ASCII smuggling, agentic AI security (OWASP Agentic Apps 2026, ASI01-ASI10). Patterns: direct injection ('ignore previous instructions'), indirect injection via documents/web pages/email the model reads, ASCII smuggling…
performing-aws-privilege-escalation-assessment
Performing authorized privilege escalation assessments in AWS environments to identify IAM misconfigurations that allow users or roles to elevate their permissions using Pacu, CloudFox, Principal Mapper, and manual IAM policy analysis techniques.
performing-cloud-penetration-testing-with-pacu
Performing authorized AWS penetration testing using Pacu, the open-source AWS exploitation framework, to enumerate IAM configurations, discover privilege escalation paths, test credential harvesting, and validate security controls through systematic attack simulation.
conducting-cloud-penetration-testing
This skill outlines methodologies for performing authorized penetration testing against AWS, Azure, and GCP cloud environments. It covers understanding the shared responsibility model for testing scope, leveraging cloud-specific attack tools like Pacu and ScoutSuite, exploiting IAM misconfigurations, testing for SSRF…
performing-kubernetes-penetration-testing
Kubernetes penetration testing systematically evaluates cluster security by simulating attacker techniques against the API server, kubelet, etcd, pods, RBAC, network policies, and secrets. Using tools.
conducting-full-scope-red-team-engagement
Plan and execute a comprehensive red team engagement covering reconnaissance through post-exploitation using MITRE ATT&CK-aligned TTPs to evaluate an organization's detection and response capabilities.