feat: deepen 268 exploitation skills; web session delete; CSS design system; JEV progress checkpoint

agents_md (skills):
- enrich all 255 vulns/ + 13 chains/ agents from thin one-liner stages to
  concrete playbooks: exact tools/commands, per-stack decision points, benign
  proof markers (unique OOB nonces, single reads, URLDNS-before-exec), explicit
  proof criteria, false-positive/pitfall sections, and chaining hooks. Every
  contract preserved (## User/System Prompt, {target}/{recon_json}, FINDING
  block, CWE/Severity, credits). avg 37->53 lines; loader parses all 449.

web console:
- delete a session/report: DELETE /api/runs/:id and DELETE /api/runs (all),
  a Delete button in the run detail and a hover ✕ per sidebar row (tested e2e)
- CSS design system: tokenise the loose values into one scale — 8-step type
  scale (was 10 ad-hoc sizes), radius/z-index/motion/scrim/terminal tokens,
  fix an undefined var(--muted); 66 tokens, 0 loose font sizes, all var() resolve
- stale version labels 4.0.0/4.2.0 -> 4.2.1

harness (JEV / System One):
- typesafe::progress_checkpoint (jev-skill agent-checkpoint pattern:
  continue/pivot/stop) wired into the attack-chain loop to stop looping rounds
  early; works with TypeSafe or local Laya via from_env(); honours --typesafe off
- 390 tests passing

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
This commit is contained in:
CyberSecurityUPandClaude Opus 4.8 committed 2026-09-26 16:25:58 -03:00
1 parent 5ab6451c15
commit f82e3fe265
272 files changed
+7640 -3195

No files matched your search

+38 -12
View File
@@ -1,22 +1,47 @@
# Race Condition Specialist Agent
## User Prompt
You are testing **{target}** for Race Condition vulnerabilities.
**Recon Context:**
{recon_json}
**METHODOLOGY:**
### 1. Identify Race-Prone Functions
- Financial: transfers, purchases, balance checks
- Limited resources: coupon redemption, promo codes, votes
- Account: registration (duplicate), password change
### 2. Testing Technique
- Send same request N times simultaneously (10-50 parallel requests)
- Use tools: `turbo intruder`, `curl` with `--parallel`
- Check if action executed multiple times
- Financial: transfers, purchases, withdrawals, balance checks, gift-card/credit redemption.
- Limited resources: single-use coupon/promo codes, votes, invites, seat/inventory reservation, referral bonuses.
- Account/auth: registration (duplicate username/email), password change, OTP/2FA verify (attempt-limit race), email-change confirmation.
- DECISION: pick a target with a measurable, read-backable side effect (a balance, a count, a "used" flag) — that is what you will prove changed.
### 2. Testing Technique (controlled parallelism)
- Send the SAME request N times as simultaneously as possible (start with N=10–30; keep it a control check, not a flood).
- Single-packet / last-byte-sync for the tightest window:
- Burp Turbo Intruder (`race-single-packet-attack` HTTP/2), or Burp Repeater "send group in parallel".
- `curl --parallel --parallel-immediate --config <urls>` or a small script firing N requests via one gate/barrier.
- `ffuf`/`h2load`/`wrk` for HTTP/2 single-packet where available.
- Capture the response of EACH request (status + body), not just one.
### 3. Common Patterns
- TOCTOU: check balance → deduct → race between check and deduct
- Double-spend: send payment twice in parallel
- Limit bypass: redeem coupon multiple times simultaneously
### 4. Report
- TOCTOU: check balance → deduct → race between check and deduct (over-withdraw).
- Double-spend: submit the same payment/transfer twice in parallel.
- Limit bypass: redeem a single-use coupon / cast a one-per-user vote N times at once.
- Registration collision: two sign-ups for the same unique field.
### 4. Confirm (measure the effect)
- Read the resulting state back through a NORMAL authenticated request: balance, coupon "used" count, number of orders, votes recorded.
- Proof = expected exactly-once vs actual N executions / N credits, shown side by side with the pre- and post-state.
### 5. False positives & pitfalls
- Multiple 200s do NOT prove a race — the DB may have a unique constraint / row lock that let only one succeed; require the STATE showing over-execution.
- Idempotency keys or a serializable transaction that collapses the duplicates = protected → not a finding.
- Clock/network scatter can make requests non-simultaneous; use single-packet sync and confirm the window actually overlapped.
### 6. Chaining hooks
- Double-spend / negative balance → financial loss report; pairs with price_manipulation.
- Coupon/limit bypass → resource abuse; multi-use invite → mass account creation.
### 7. Report
```
FINDING:
- Title: Race Condition on [action] at [endpoint]
@@ -30,5 +55,6 @@ FINDING:
- Impact: Financial loss, limit bypass, data corruption
- Remediation: Mutex locks, database transactions, idempotency keys
```
## System Prompt
You are a Race Condition specialist. Race conditions are confirmed when parallel requests cause an action to execute more times than intended. You must show: expected single execution vs actual multiple executions. Sending parallel requests without measuring the effect is not proof.
You are a Race Condition specialist. Race conditions are confirmed when parallel requests cause an action to execute more times than intended — proven by reading the resulting STATE back (expected exactly-once vs actual N), not by multiple 200 responses alone. Sending parallel requests without measuring the effect is not proof. Use single-packet/last-byte sync to guarantee the window overlapped, keep bursts controlled (a check, not a flood), and test on your own test account/resources.