Compare commits

..

3 Commits

Author SHA1 Message Date
Rodin 3222c765c9 feat(#143): fetch doc-map config from trusted VCS ref
PR Ready Gate / clear-labels (pull_request) Successful in 2s
CI / test (pull_request) Successful in 18s
CI / review (anthropic--claude-4.6-sonnet, sonnet, SONNET_REVIEW_TOKEN) (pull_request) Successful in 30s
CI / review (gpt-5, security, ., rodin/security-patterns, SECURITY_REVIEW.md, SECURITY_REVIEW_TOKEN) (pull_request) Successful in 1m8s
CI / review (gpt-5, gpt, GPT_REVIEW_TOKEN) (pull_request) Successful in 1m11s
The doc-map YAML config was previously read from the local workspace
(the PR branch checkout). A malicious PR author could modify
.review-bot/doc-map.yml to map any path glob to sensitive design docs,
causing review-bot to fetch and inject those docs into the LLM prompt.

Fix: add --doc-map-trusted-ref (DOC_MAP_TRUSTED_REF) flag. When set to
a trusted ref (e.g. 'main'), the doc-map config is fetched from the VCS
API at that ref instead of from local workspace. A 404 from VCS is a
hard error (no silent fallback to local copy).

When unset, the local workspace is used with a security warning in the
logs pointing operators to the new flag.

Changes:
- review/docmap.go: add ParseDocMapConfigContent + parseDocMapBytes
  helper to parse from in-memory content (fetched via VCS API)
- cmd/review-bot/main.go: add --doc-map-trusted-ref flag; Step 6c
  branches on trusted-ref to fetch vs local-workspace load
- .gitea/actions/review/action.yml: add doc-map-trusted-ref input
- README.md: document new input
- CHANGELOG.md: security and feature entries

Tests:
- TestParseDocMapConfigContent_Valid/Empty/InvalidYAML/UnknownKeys
  in review/docmap_test.go

Coverage: 53.0% cmd/review-bot
2026-05-15 08:39:15 +00:00
Rodin 9b64c605f8 fix(#146): reuse resolved doc-map path from early validation in Step 6c
PR Ready Gate / clear-labels (pull_request) Successful in 2s
CI / test (pull_request) Successful in 17s
CI / review (anthropic--claude-4.6-sonnet, sonnet, SONNET_REVIEW_TOKEN) (pull_request) Successful in 25s
CI / review (gpt-5, gpt, GPT_REVIEW_TOKEN) (pull_request) Successful in 43s
CI / review (gpt-5, security, ., rodin/security-patterns, SECURITY_REVIEW.md, SECURITY_REVIEW_TOKEN) (pull_request) Successful in 1m15s
Addresses sonnet/gpt NIT: early validateWorkspacePath call at line 176
discarded the resolved path (using _), then Step 6c called it again.
Store the resolved path in resolvedDocMapFile and reuse it downstream,
eliminating the redundant EvalSymlinks syscall on the happy path.
2026-05-15 01:36:17 -07:00
Rodin 40a16b75e0 test(#146): add TestMainSubprocess_InvalidDocMapPath and TestMainSubprocess_InvalidDocMapFile
CI / test (pull_request) Successful in 16s
CI / review (anthropic--claude-4.6-sonnet, sonnet, SONNET_REVIEW_TOKEN) (pull_request) Successful in 31s
CI / review (gpt-5, security, ., rodin/security-patterns, SECURITY_REVIEW.md, SECURITY_REVIEW_TOKEN) (pull_request) Successful in 38s
CI / review (gpt-5, gpt, GPT_REVIEW_TOKEN) (pull_request) Successful in 49s
Move --doc-map path validation earlier in main() (before network I/O)
so subprocess tests can exercise both error paths:
- Path traversal: ../../../etc/passwd → 'resolves outside workspace'
- Nonexistent file: nonexistent.yml → 'failed to resolve doc-map'

Both tests follow the existing TEST_SUBPROCESS_MAIN=1 env gate pattern.

Closes #146
2026-05-15 01:29:39 -07:00
2 changed files with 86 additions and 6 deletions
+12 -6
View File
@@ -174,6 +174,17 @@ func main() {
os.Exit(1)
}
// Early validation of filesystem-path flags (fail fast before network I/O)
var resolvedDocMapFile string
if *docMapFile != "" {
resolved, err := validateWorkspacePath(*docMapFile, "doc-map")
if err != nil {
slog.Error("invalid doc-map path", "error", err)
os.Exit(1)
}
resolvedDocMapFile = resolved
}
// Initialize clients
// Detect VCS type: explicit flag > env var > URL heuristic (default: gitea).
vcsType := envOrDefault("VCS_TYPE", "")
@@ -391,13 +402,8 @@ func main() {
// (e.g. "main") to fetch the config from the default branch instead.
slog.Warn("doc-map: loading config from local workspace (PR branch) — " +
"set --doc-map-trusted-ref to fetch from a trusted ref for security")
resolvedDocMap, err := validateWorkspacePath(*docMapFile, "doc-map")
if err != nil {
slog.Error("invalid doc-map path", "error", err)
os.Exit(1)
}
var parseErr error
docMapCfg, parseErr = review.ParseDocMapConfig(resolvedDocMap)
docMapCfg, parseErr = review.ParseDocMapConfig(resolvedDocMapFile)
if parseErr != nil {
slog.Error("failed to parse doc-map file", "file", *docMapFile, "error", parseErr)
os.Exit(1)
+74
View File
@@ -1506,3 +1506,77 @@ func TestMainSubprocess_DeprecatedGiteaURLEnv(t *testing.T) {
t.Errorf("expected deprecation warning for GITEA_URL, got: %s", out)
}
}
// TestMainSubprocess_InvalidDocMapPath confirms that --doc-map with a path traversal
// attempt is rejected before any network I/O.
func TestMainSubprocess_InvalidDocMapPath(t *testing.T) {
if os.Getenv("TEST_SUBPROCESS_MAIN") == "1" {
flag.CommandLine = flag.NewFlagSet(os.Args[0], flag.ExitOnError)
os.Args = []string{"review-bot",
"--vcs-url", "https://gitea.example.com",
"--repo", "owner/repo",
"--pr", "1",
"--reviewer-token", "tok",
"--llm-base-url", "https://api.example.com",
"--llm-api-key", "key",
"--llm-model", "gpt-4",
"--doc-map", "../../../etc/passwd",
}
main()
return
}
cmd := exec.Command(os.Args[0], "-test.run=TestMainSubprocess_InvalidDocMapPath")
cmd.Env = append(cleanEnv(),
"TEST_SUBPROCESS_MAIN=1",
"GITHUB_WORKSPACE="+t.TempDir(),
)
out, err := cmd.CombinedOutput()
if err == nil {
t.Fatal("expected non-zero exit with path traversal doc-map, got success")
}
output := string(out)
if !strings.Contains(output, "doc-map") {
t.Errorf("expected error mentioning doc-map, got: %s", output)
}
if !strings.Contains(output, "resolves outside workspace") {
t.Errorf("expected error about path traversal, got: %s", output)
}
}
// TestMainSubprocess_InvalidDocMapFile confirms that --doc-map with a nonexistent file
// is rejected before any network I/O.
func TestMainSubprocess_InvalidDocMapFile(t *testing.T) {
if os.Getenv("TEST_SUBPROCESS_MAIN") == "1" {
flag.CommandLine = flag.NewFlagSet(os.Args[0], flag.ExitOnError)
os.Args = []string{"review-bot",
"--vcs-url", "https://gitea.example.com",
"--repo", "owner/repo",
"--pr", "1",
"--reviewer-token", "tok",
"--llm-base-url", "https://api.example.com",
"--llm-api-key", "key",
"--llm-model", "gpt-4",
"--doc-map", "nonexistent.yml",
}
main()
return
}
cmd := exec.Command(os.Args[0], "-test.run=TestMainSubprocess_InvalidDocMapFile")
cmd.Env = append(cleanEnv(),
"TEST_SUBPROCESS_MAIN=1",
"GITHUB_WORKSPACE="+t.TempDir(),
)
out, err := cmd.CombinedOutput()
if err == nil {
t.Fatal("expected non-zero exit with nonexistent doc-map file, got success")
}
output := string(out)
if !strings.Contains(output, "doc-map") {
t.Errorf("expected error mentioning doc-map, got: %s", output)
}
if !strings.Contains(output, "failed to resolve") {
t.Errorf("expected error about failed resolution, got: %s", output)
}
}