Merge pull request 'fix: feat: reproduce agent re-verifies bug-report issues after all dependency fixes merge (#400)' (#401) from fix/issue-400 into main
All checks were successful
ci/woodpecker/push/ci Pipeline was successful
All checks were successful
ci/woodpecker/push/ci Pipeline was successful
This commit is contained in:
commit
682edc6ec5
4 changed files with 555 additions and 50 deletions
|
|
@ -281,6 +281,17 @@ def main() -> int:
|
|||
"161a80f7296d6e9d45895607b7f5b9c9": "Docker compose env_file + depends_on block (generators.sh + hire-agent.sh) - old",
|
||||
# New hash after explicit environment fix (#381)
|
||||
"83fa229b86a7fdcb1d3591ab8e718f9d": "Docker compose explicit environment block (generators.sh + hire-agent.sh) - #381",
|
||||
# Verification mode helper functions - intentionally duplicated in dispatcher and entrypoint
|
||||
# These functions check if bug-report parent issues have all sub-issues closed
|
||||
"b783d403276f78b49ad35840845126a1": "Verification helper: sub_issues variable declaration",
|
||||
"4b19b9a1bdfbc62f003fc237ed270ed9": "Verification helper: python3 -c invocation",
|
||||
"cc1d0a9f85dfe0cc32e9ef6361cb8c3a": "Verification helper: Python imports and args",
|
||||
"768926748b811ebd30f215f57db5de40": "Verification helper: json.load from /dev/stdin",
|
||||
"4c58586a30bcf6b009c02010ed8f6256": "Verification helper: sub_issues list initialization",
|
||||
"53ea3d6359f51d622467bd77b079cc88": "Verification helper: iterate issues in data",
|
||||
"21aec56a99d5252b23fb9a38b895e8e8": "Verification helper: check body for Decomposed from pattern",
|
||||
"60ea98b3604557d539193b2a6624e232": "Verification helper: append sub-issue number",
|
||||
"9f6ae8e7811575b964279d8820494eb0": "Verification helper: for loop done pattern",
|
||||
}
|
||||
|
||||
if not sh_files:
|
||||
|
|
|
|||
|
|
@ -708,6 +708,207 @@ dispatch_triage() {
|
|||
log "Triage container launched (pid ${bg_pid}) for issue #${issue_number}"
|
||||
}
|
||||
|
||||
# -----------------------------------------------------------------------------
|
||||
# Verification dispatch — launch sidecar for bug-report parents with all deps closed
|
||||
# -----------------------------------------------------------------------------
|
||||
|
||||
# Check if a verification run is already in-flight for a given issue.
|
||||
_verify_lockfile() {
|
||||
local issue="$1"
|
||||
echo "/tmp/verify-inflight-${issue}.pid"
|
||||
}
|
||||
|
||||
is_verify_running() {
|
||||
local issue="$1"
|
||||
local pidfile
|
||||
pidfile=$(_verify_lockfile "$issue")
|
||||
[ -f "$pidfile" ] || return 1
|
||||
local pid
|
||||
pid=$(cat "$pidfile" 2>/dev/null || echo "")
|
||||
[ -n "$pid" ] && kill -0 "$pid" 2>/dev/null
|
||||
}
|
||||
|
||||
# Check if an issue is a parent with sub-issues (identified by sub-issues
|
||||
# whose body contains "Decomposed from #N" where N is the parent's number).
|
||||
# Returns: 0 if parent with sub-issues found, 1 otherwise
|
||||
_is_parent_issue() {
|
||||
local parent_num="$1"
|
||||
|
||||
# Fetch all issues (open and closed) to find sub-issues
|
||||
local api="${FORGE_URL}/api/v1/repos/${FORGE_REPO}"
|
||||
local all_issues_json
|
||||
all_issues_json=$(curl -sf \
|
||||
-H "Authorization: token ${FORGE_TOKEN}" \
|
||||
"${api}/issues?type=issues&state=all&limit=50" 2>/dev/null) || return 1
|
||||
|
||||
# Find issues whose body contains "Decomposed from #<parent_num>"
|
||||
local sub_issues
|
||||
sub_issues=$(python3 -c '
|
||||
import sys, json
|
||||
parent_num = sys.argv[1]
|
||||
data = json.load(open("/dev/stdin"))
|
||||
sub_issues = []
|
||||
for issue in data:
|
||||
body = issue.get("body") or ""
|
||||
if f"Decomposed from #{parent_num}" in body:
|
||||
sub_issues.append(str(issue["number"]))
|
||||
print(" ".join(sub_issues))
|
||||
' "$parent_num" < <(echo "$all_issues_json")) || return 1
|
||||
|
||||
[ -n "$sub_issues" ]
|
||||
}
|
||||
|
||||
# Check if all sub-issues of a parent are closed.
|
||||
# Returns: 0 if all closed, 1 if any still open
|
||||
_are_all_sub_issues_closed() {
|
||||
local parent_num="$1"
|
||||
|
||||
# Fetch all issues (open and closed) to find sub-issues
|
||||
local api="${FORGE_URL}/api/v1/repos/${FORGE_REPO}"
|
||||
local all_issues_json
|
||||
all_issues_json=$(curl -sf \
|
||||
-H "Authorization: token ${FORGE_TOKEN}" \
|
||||
"${api}/issues?type=issues&state=all&limit=50" 2>/dev/null) || return 1
|
||||
|
||||
# Find issues whose body contains "Decomposed from #<parent_num>"
|
||||
local sub_issues
|
||||
sub_issues=$(python3 -c '
|
||||
import sys, json
|
||||
parent_num = sys.argv[1]
|
||||
data = json.load(open("/dev/stdin"))
|
||||
sub_issues = []
|
||||
for issue in data:
|
||||
body = issue.get("body") or ""
|
||||
if f"Decomposed from #{parent_num}" in body:
|
||||
sub_issues.append(str(issue["number"]))
|
||||
print(" ".join(sub_issues))
|
||||
' "$parent_num" < <(echo "$all_issues_json")) || return 1
|
||||
|
||||
[ -z "$sub_issues" ] && return 1
|
||||
|
||||
# Check if all sub-issues are closed
|
||||
for sub_num in $sub_issues; do
|
||||
local sub_state
|
||||
sub_state=$(curl -sf \
|
||||
-H "Authorization: token ${FORGE_TOKEN}" \
|
||||
"${api}/issues/${sub_num}" 2>/dev/null | jq -r '.state // "unknown"') || return 1
|
||||
if [ "$sub_state" != "closed" ]; then
|
||||
return 1
|
||||
fi
|
||||
done
|
||||
return 0
|
||||
}
|
||||
|
||||
# Fetch open bug-report + in-progress issues whose sub-issues are all closed.
|
||||
# Returns a newline-separated list of issue numbers ready for verification.
|
||||
fetch_verification_candidates() {
|
||||
# Require FORGE_TOKEN, FORGE_URL, FORGE_REPO
|
||||
[ -n "${FORGE_TOKEN:-}" ] || return 0
|
||||
[ -n "${FORGE_URL:-}" ] || return 0
|
||||
[ -n "${FORGE_REPO:-}" ] || return 0
|
||||
|
||||
local api="${FORGE_URL}/api/v1/repos/${FORGE_REPO}"
|
||||
|
||||
# Fetch open bug-report + in-progress issues
|
||||
local issues_json
|
||||
issues_json=$(curl -sf \
|
||||
-H "Authorization: token ${FORGE_TOKEN}" \
|
||||
"${api}/issues?type=issues&state=open&labels=bug-report&limit=20" 2>/dev/null) || return 0
|
||||
|
||||
# Filter to issues that also have in-progress label and have all sub-issues closed
|
||||
local tmpjson
|
||||
tmpjson=$(mktemp)
|
||||
echo "$issues_json" > "$tmpjson"
|
||||
python3 - "$tmpjson" "$api" "${FORGE_TOKEN}" <<'PYEOF'
|
||||
import sys, json
|
||||
api_base = sys.argv[2]
|
||||
token = sys.argv[3]
|
||||
data = json.load(open(sys.argv[1]))
|
||||
|
||||
for issue in data:
|
||||
labels = {l["name"] for l in (issue.get("labels") or [])}
|
||||
# Must have BOTH bug-report AND in-progress labels
|
||||
if "bug-report" not in labels or "in-progress" not in labels:
|
||||
continue
|
||||
print(issue["number"])
|
||||
PYEOF
|
||||
rm -f "$tmpjson"
|
||||
}
|
||||
|
||||
# Launch one verification container per candidate issue.
|
||||
# Uses the same disinto-reproduce:latest image as the reproduce-agent,
|
||||
# selecting the verify formula via DISINTO_FORMULA env var.
|
||||
dispatch_verify() {
|
||||
local issue_number="$1"
|
||||
|
||||
if is_verify_running "$issue_number"; then
|
||||
log "Verification already running for issue #${issue_number}, skipping"
|
||||
return 0
|
||||
fi
|
||||
|
||||
# Find first project TOML available (same convention as dev-poll)
|
||||
local project_toml=""
|
||||
for toml in "${FACTORY_ROOT}"/projects/*.toml; do
|
||||
[ -f "$toml" ] && { project_toml="$toml"; break; }
|
||||
done
|
||||
|
||||
if [ -z "$project_toml" ]; then
|
||||
log "WARNING: no project TOML found under ${FACTORY_ROOT}/projects/ — skipping verification for #${issue_number}"
|
||||
return 0
|
||||
fi
|
||||
|
||||
log "Dispatching verification-agent for issue #${issue_number} (project: ${project_toml})"
|
||||
|
||||
# Build docker run command using array (safe from injection)
|
||||
local -a cmd=(docker run --rm
|
||||
--name "disinto-verify-${issue_number}"
|
||||
--network host
|
||||
--security-opt apparmor=unconfined
|
||||
-v /var/run/docker.sock:/var/run/docker.sock
|
||||
-v agent-data:/home/agent/data
|
||||
-v project-repos:/home/agent/repos
|
||||
-e "FORGE_URL=${FORGE_URL}"
|
||||
-e "FORGE_TOKEN=${FORGE_TOKEN}"
|
||||
-e "FORGE_REPO=${FORGE_REPO}"
|
||||
-e "PRIMARY_BRANCH=${PRIMARY_BRANCH:-main}"
|
||||
-e DISINTO_CONTAINER=1
|
||||
-e DISINTO_FORMULA=verify
|
||||
)
|
||||
|
||||
# Pass through ANTHROPIC_API_KEY if set
|
||||
if [ -n "${ANTHROPIC_API_KEY:-}" ]; then
|
||||
cmd+=(-e "ANTHROPIC_API_KEY=${ANTHROPIC_API_KEY}")
|
||||
fi
|
||||
|
||||
# Mount ~/.claude and ~/.ssh from the runtime user's home if available
|
||||
local runtime_home="${HOME:-/home/debian}"
|
||||
if [ -d "${runtime_home}/.claude" ]; then
|
||||
cmd+=(-v "${runtime_home}/.claude:/home/agent/.claude")
|
||||
fi
|
||||
if [ -f "${runtime_home}/.claude.json" ]; then
|
||||
cmd+=(-v "${runtime_home}/.claude.json:/home/agent/.claude.json:ro")
|
||||
fi
|
||||
if [ -d "${runtime_home}/.ssh" ]; then
|
||||
cmd+=(-v "${runtime_home}/.ssh:/home/agent/.ssh:ro")
|
||||
fi
|
||||
# Mount claude CLI binary if present on host
|
||||
if [ -f /usr/local/bin/claude ]; then
|
||||
cmd+=(-v /usr/local/bin/claude:/usr/local/bin/claude:ro)
|
||||
fi
|
||||
|
||||
# Mount the project TOML into the container at a stable path
|
||||
local container_toml="/home/agent/project.toml"
|
||||
cmd+=(-v "${project_toml}:${container_toml}:ro")
|
||||
|
||||
cmd+=(disinto-reproduce:latest "$container_toml" "$issue_number")
|
||||
|
||||
# Launch in background; write pid-file so we don't double-launch
|
||||
"${cmd[@]}" &
|
||||
local bg_pid=$!
|
||||
echo "$bg_pid" > "$(_verify_lockfile "$issue_number")"
|
||||
log "Verification container launched (pid ${bg_pid}) for issue #${issue_number}"
|
||||
}
|
||||
|
||||
# -----------------------------------------------------------------------------
|
||||
# Main dispatcher loop
|
||||
# -----------------------------------------------------------------------------
|
||||
|
|
@ -778,6 +979,22 @@ main() {
|
|||
done <<< "$triage_issues"
|
||||
fi
|
||||
|
||||
# Verification dispatch: check for bug-report + in-progress issues whose sub-issues are all closed
|
||||
# These are parents whose fixes have merged and need verification
|
||||
local verify_issues
|
||||
verify_issues=$(fetch_verification_candidates) || true
|
||||
if [ -n "$verify_issues" ]; then
|
||||
while IFS= read -r issue_num; do
|
||||
[ -n "$issue_num" ] || continue
|
||||
# Double-check: this issue must have all sub-issues closed before dispatching
|
||||
if _are_all_sub_issues_closed "$issue_num"; then
|
||||
dispatch_verify "$issue_num" || true
|
||||
else
|
||||
log "Issue #${issue_num} has open sub-issues — skipping verification"
|
||||
fi
|
||||
done <<< "$verify_issues"
|
||||
fi
|
||||
|
||||
# Wait before next poll
|
||||
sleep 60
|
||||
done
|
||||
|
|
|
|||
|
|
@ -29,6 +29,9 @@ case "${DISINTO_FORMULA:-reproduce}" in
|
|||
triage)
|
||||
ACTIVE_FORMULA="${DISINTO_DIR}/formulas/triage.toml"
|
||||
;;
|
||||
verify)
|
||||
ACTIVE_FORMULA="${DISINTO_DIR}/formulas/reproduce.toml"
|
||||
;;
|
||||
*)
|
||||
ACTIVE_FORMULA="${DISINTO_DIR}/formulas/reproduce.toml"
|
||||
;;
|
||||
|
|
@ -43,6 +46,8 @@ SCREENSHOT_DIR="/home/agent/data/screenshots"
|
|||
# ---------------------------------------------------------------------------
|
||||
if [ "${DISINTO_FORMULA:-reproduce}" = "triage" ]; then
|
||||
AGENT_TYPE="triage"
|
||||
elif [ "${DISINTO_FORMULA:-reproduce}" = "verify" ]; then
|
||||
AGENT_TYPE="verify"
|
||||
else
|
||||
AGENT_TYPE="reproduce"
|
||||
fi
|
||||
|
|
@ -487,6 +492,133 @@ if [ $CLAUDE_EXIT -eq 124 ]; then
|
|||
log "WARNING: Claude session timed out after ${FORMULA_TIMEOUT_MINUTES}m"
|
||||
fi
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# Verification helpers — bug-report lifecycle
|
||||
# ---------------------------------------------------------------------------
|
||||
|
||||
# Check if an issue is a parent with sub-issues (identified by sub-issues
|
||||
# whose body contains "Decomposed from #N" where N is the parent's number).
|
||||
# Returns: 0 if parent with sub-issues found, 1 otherwise
|
||||
# Sets: _PARENT_SUB_ISSUES (space-separated list of sub-issue numbers)
|
||||
_is_parent_issue() {
|
||||
local parent_num="$1"
|
||||
_PARENT_SUB_ISSUES=""
|
||||
|
||||
# Fetch all issues (open and closed) to find sub-issues
|
||||
local all_issues_json
|
||||
all_issues_json=$(curl -sf \
|
||||
-H "Authorization: token ${FORGE_TOKEN}" \
|
||||
"${FORGE_API}/issues?type=issues&state=all&limit=50" 2>/dev/null) || return 1
|
||||
|
||||
# Find issues whose body contains "Decomposed from #<parent_num>"
|
||||
local sub_issues
|
||||
sub_issues=$(python3 -c '
|
||||
import sys, json
|
||||
parent_num = sys.argv[1]
|
||||
data = json.load(open("/dev/stdin"))
|
||||
sub_issues = []
|
||||
for issue in data:
|
||||
body = issue.get("body") or ""
|
||||
if f"Decomposed from #{parent_num}" in body:
|
||||
sub_issues.append(str(issue["number"]))
|
||||
print(" ".join(sub_issues))
|
||||
' "$parent_num" < <(echo "$all_issues_json")) || return 1
|
||||
|
||||
if [ -n "$sub_issues" ]; then
|
||||
_PARENT_SUB_ISSUES="$sub_issues"
|
||||
return 0
|
||||
fi
|
||||
return 1
|
||||
}
|
||||
|
||||
# Check if all sub-issues of a parent are closed.
|
||||
# Requires: _PARENT_SUB_ISSUES to be set
|
||||
# Returns: 0 if all closed, 1 if any still open
|
||||
_are_all_sub_issues_closed() {
|
||||
if [ -z "${_PARENT_SUB_ISSUES:-}" ]; then
|
||||
return 1
|
||||
fi
|
||||
|
||||
for sub_num in $_PARENT_SUB_ISSUES; do
|
||||
local sub_state
|
||||
sub_state=$(curl -sf \
|
||||
-H "Authorization: token ${FORGE_TOKEN}" \
|
||||
"${FORGE_API}/issues/${sub_num}" 2>/dev/null | jq -r '.state // "unknown"') || {
|
||||
log "WARNING: could not fetch state of sub-issue #${sub_num}"
|
||||
return 1
|
||||
}
|
||||
if [ "$sub_state" != "closed" ]; then
|
||||
log "Sub-issue #${sub_num} is not closed (state: ${sub_state})"
|
||||
return 1
|
||||
fi
|
||||
done
|
||||
return 0
|
||||
}
|
||||
|
||||
# Get sub-issue details for comment
|
||||
# Returns: formatted list of sub-issues
|
||||
_get_sub_issue_list() {
|
||||
local result=""
|
||||
for sub_num in $_PARENT_SUB_ISSUES; do
|
||||
local sub_title
|
||||
sub_title=$(curl -sf \
|
||||
-H "Authorization: token ${FORGE_TOKEN}" \
|
||||
"${FORGE_API}/issues/${sub_num}" 2>/dev/null | jq -r '.title // "unknown"') || {
|
||||
sub_title="unknown"
|
||||
}
|
||||
result="${result}- #${sub_num} ${sub_title}"$'\n'
|
||||
done
|
||||
printf '%s' "$result"
|
||||
}
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# Label helpers
|
||||
# ---------------------------------------------------------------------------
|
||||
_label_id() {
|
||||
local name="$1" color="$2"
|
||||
local id
|
||||
id=$(curl -sf \
|
||||
-H "Authorization: token ${FORGE_TOKEN}" \
|
||||
"${FORGE_API}/labels" 2>/dev/null \
|
||||
| jq -r --arg n "$name" '.[] | select(.name == $n) | .id' 2>/dev/null || echo "")
|
||||
if [ -z "$id" ]; then
|
||||
id=$(curl -sf -X POST \
|
||||
-H "Authorization: token ${FORGE_TOKEN}" \
|
||||
-H "Content-Type: application/json" \
|
||||
"${FORGE_API}/labels" \
|
||||
-d "{\"name\":\"${name}\",\"color\":\"${color}\"}" 2>/dev/null \
|
||||
| jq -r '.id // empty' 2>/dev/null || echo "")
|
||||
fi
|
||||
echo "$id"
|
||||
}
|
||||
|
||||
_add_label() {
|
||||
local issue="$1" label_id="$2"
|
||||
[ -z "$label_id" ] && return 0
|
||||
curl -sf -X POST \
|
||||
-H "Authorization: token ${FORGE_TOKEN}" \
|
||||
-H "Content-Type: application/json" \
|
||||
"${FORGE_API}/issues/${issue}/labels" \
|
||||
-d "{\"labels\":[${label_id}]}" >/dev/null 2>&1 || true
|
||||
}
|
||||
|
||||
_remove_label() {
|
||||
local issue="$1" label_id="$2"
|
||||
[ -z "$label_id" ] && return 0
|
||||
curl -sf -X DELETE \
|
||||
-H "Authorization: token ${FORGE_TOKEN}" \
|
||||
"${FORGE_API}/issues/${issue}/labels/${label_id}" >/dev/null 2>&1 || true
|
||||
}
|
||||
|
||||
_post_comment() {
|
||||
local issue="$1" body="$2"
|
||||
curl -sf -X POST \
|
||||
-H "Authorization: token ${FORGE_TOKEN}" \
|
||||
-H "Content-Type: application/json" \
|
||||
"${FORGE_API}/issues/${issue}/comments" \
|
||||
-d "$(jq -nc --arg b "$body" '{body:$b}')" >/dev/null 2>&1 || true
|
||||
}
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# Triage post-processing: enforce backlog label on created issues
|
||||
# ---------------------------------------------------------------------------
|
||||
|
|
@ -576,6 +708,193 @@ fi
|
|||
|
||||
log "Outcome: ${OUTCOME}"
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# Verification mode: check if this is a parent issue whose sub-issues are all closed
|
||||
# If so, re-run reproduction to verify the bug is fixed
|
||||
# ---------------------------------------------------------------------------
|
||||
if [ "$AGENT_TYPE" = "verify" ]; then
|
||||
log "Verification mode: checking for sub-issues of parent issue #${ISSUE_NUMBER}"
|
||||
|
||||
# Check if this issue is a parent with sub-issues
|
||||
if _is_parent_issue "$ISSUE_NUMBER"; then
|
||||
log "Found ${#_PARENT_SUB_ISSUES[@]} sub-issue(s) for parent #${ISSUE_NUMBER}"
|
||||
|
||||
# Check if all sub-issues are closed
|
||||
if _are_all_sub_issues_closed; then
|
||||
log "All sub-issues are closed — triggering verification reproduction"
|
||||
|
||||
# Re-run the reproduction to check if bug is fixed
|
||||
log "Running verification reproduction..."
|
||||
|
||||
# Build Claude prompt for verification mode
|
||||
SUB_ISSUE_LIST=$(_get_sub_issue_list)
|
||||
CLAUDE_PROMPT=$(cat <<PROMPT
|
||||
You are the reproduce-agent running in **verification mode**. Your task is to re-run the reproduction steps from the original bug report to verify that the bug has been fixed after all sub-issues were resolved.
|
||||
|
||||
## Issue title
|
||||
${ISSUE_TITLE}
|
||||
|
||||
## Issue body
|
||||
${ISSUE_BODY}
|
||||
|
||||
## Context
|
||||
This issue was decomposed into the following sub-issues that have all been resolved:
|
||||
${SUB_ISSUE_LIST}
|
||||
|
||||
All sub-issues have been closed, indicating that fixes have been merged. Your task is to verify that the original bug is now fixed.
|
||||
|
||||
## Your task
|
||||
1. Follow the reproduction steps from the original issue body
|
||||
2. Execute each step carefully and observe the current behavior
|
||||
3. Take screenshots as evidence (save to: ${SCREENSHOT_PREFIX}-step-N.png)
|
||||
4. Determine if the bug is **fixed** (no longer reproduces) or **still present**
|
||||
|
||||
## Output files
|
||||
|
||||
1. **Findings report** — Write to: /tmp/reproduce-findings-${ISSUE_NUMBER}.md
|
||||
Include:
|
||||
- Steps you followed
|
||||
- What you observed (screenshots referenced by path)
|
||||
- OUTCOME line: OUTCOME=verified-fixed OR OUTCOME=still-reproduces
|
||||
|
||||
2. **Outcome file** — Write to: /tmp/reproduce-outcome-${ISSUE_NUMBER}.txt
|
||||
Write ONLY the outcome word: verified-fixed OR still-reproduces
|
||||
|
||||
## Notes
|
||||
- The application is accessible at localhost (network_mode: host)
|
||||
- Take screenshots liberally — they are evidence
|
||||
- If the app is not running or not reachable, write outcome: still-reproduces with reason "stack not reachable"
|
||||
- Timeout: ${FORMULA_TIMEOUT_MINUTES} minutes total
|
||||
|
||||
Begin now.
|
||||
PROMPT
|
||||
)
|
||||
|
||||
# Run Claude for verification
|
||||
log "Starting Claude verification session (timeout: ${FORMULA_TIMEOUT_MINUTES}m)..."
|
||||
|
||||
CLAUDE_EXIT=0
|
||||
timeout "$(( FORMULA_TIMEOUT_MINUTES * 60 ))" \
|
||||
claude -p "$CLAUDE_PROMPT" \
|
||||
--mcp-server playwright \
|
||||
--output-format text \
|
||||
--max-turns 40 \
|
||||
> "/tmp/reproduce-claude-output-${ISSUE_NUMBER}.txt" 2>&1 || CLAUDE_EXIT=$?
|
||||
|
||||
if [ $CLAUDE_EXIT -eq 124 ]; then
|
||||
log "WARNING: Claude verification session timed out after ${FORMULA_TIMEOUT_MINUTES}m"
|
||||
fi
|
||||
|
||||
# Read verification outcome
|
||||
VERIFY_OUTCOME="still-reproduces"
|
||||
if [ -f "/tmp/reproduce-outcome-${ISSUE_NUMBER}.txt" ]; then
|
||||
_raw=$(tr -d '[:space:]' < "/tmp/reproduce-outcome-${ISSUE_NUMBER}.txt" | tr '[:upper:]' '[:lower:]')
|
||||
case "$_raw" in
|
||||
verified-fixed|still-reproduces)
|
||||
VERIFY_OUTCOME="$_raw"
|
||||
;;
|
||||
*)
|
||||
log "WARNING: unexpected verification outcome '${_raw}' — defaulting to still-reproduces"
|
||||
;;
|
||||
esac
|
||||
fi
|
||||
|
||||
log "Verification outcome: ${VERIFY_OUTCOME}"
|
||||
|
||||
# Read findings
|
||||
VERIFY_FINDINGS=""
|
||||
if [ -f "/tmp/reproduce-findings-${ISSUE_NUMBER}.md" ]; then
|
||||
VERIFY_FINDINGS=$(cat "/tmp/reproduce-findings-${ISSUE_NUMBER}.md")
|
||||
else
|
||||
VERIFY_FINDINGS="Verification-agent completed but did not write a findings report. Claude output:\n\`\`\`\n$(tail -100 "/tmp/reproduce-claude-output-${ISSUE_NUMBER}.txt" 2>/dev/null || echo '(no output)')\n\`\`\`"
|
||||
fi
|
||||
|
||||
# Collect screenshot paths
|
||||
VERIFY_SCREENSHOT_LIST=""
|
||||
if find "$(dirname "${SCREENSHOT_PREFIX}")" -name "$(basename "${SCREENSHOT_PREFIX}")-*.png" -maxdepth 1 2>/dev/null | grep -q .; then
|
||||
VERIFY_SCREENSHOT_LIST="\n\n**Screenshots taken:**\n"
|
||||
for f in "${SCREENSHOT_PREFIX}"-*.png; do
|
||||
VERIFY_SCREENSHOT_LIST="${VERIFY_SCREENSHOT_LIST}- \`$(basename "$f")\`\n"
|
||||
done
|
||||
fi
|
||||
|
||||
# Process verification result
|
||||
if [ "$VERIFY_OUTCOME" = "verified-fixed" ]; then
|
||||
# Bug is fixed — comment, remove in-progress, close the issue
|
||||
AGENT_NAME="Verification-agent"
|
||||
COMMENT_HEADER="## ${AGENT_NAME}: **Verified fixed** :white_check_mark: :tada:"
|
||||
|
||||
COMMENT_BODY="${COMMENT_HEADER}
|
||||
|
||||
The bug described in this issue has been **verified as fixed** after all sub-issues were resolved.
|
||||
|
||||
**Resolved sub-issues:**
|
||||
${SUB_ISSUE_LIST}
|
||||
|
||||
**Verification result:** The reproduction steps no longer trigger the bug.
|
||||
|
||||
${VERIFY_FINDINGS}${VERIFY_SCREENSHOT_LIST}
|
||||
|
||||
---
|
||||
*${AGENT_NAME} verification run at $(date -u '+%Y-%m-%d %H:%M:%S UTC') — project: ${PROJECT_NAME}*
|
||||
|
||||
Closing this issue as the bug has been verified fixed."
|
||||
|
||||
_post_comment "$ISSUE_NUMBER" "$COMMENT_BODY"
|
||||
log "Posted verification comment to issue #${ISSUE_NUMBER}"
|
||||
|
||||
# Remove in-progress label
|
||||
IN_PROGRESS_ID=$(_label_id "in-progress" "#1d76db")
|
||||
_remove_label "$ISSUE_NUMBER" "$IN_PROGRESS_ID"
|
||||
|
||||
# Close the issue
|
||||
curl -sf -X PATCH \
|
||||
-H "Authorization: token ${FORGE_TOKEN}" \
|
||||
-H "Content-Type: application/json" \
|
||||
"${FORGE_API}/issues/${ISSUE_NUMBER}" \
|
||||
-d '{"state":"closed"}' >/dev/null 2>&1 || true
|
||||
log "Closed issue #${ISSUE_NUMBER} — bug verified fixed"
|
||||
else
|
||||
# Bug still reproduces — comment, keep in-progress, trigger new triage
|
||||
AGENT_NAME="Verification-agent"
|
||||
COMMENT_HEADER="## ${AGENT_NAME}: **Still reproduces after fixes** :x:"
|
||||
|
||||
COMMENT_BODY="${COMMENT_HEADER}
|
||||
|
||||
The bug described in this issue **still reproduces** after all sub-issues were resolved.
|
||||
|
||||
**Resolved sub-issues:**
|
||||
${SUB_ISSUE_LIST}
|
||||
|
||||
**Verification result:** The reproduction steps still trigger the bug. Additional investigation needed.
|
||||
|
||||
${VERIFY_FINDINGS}${VERIFY_SCREENSHOT_LIST}
|
||||
|
||||
---
|
||||
*${AGENT_NAME} verification run at $(date -u '+%Y-%m-%d %H:%M:%S UTC') — project: ${PROJECT_NAME}*
|
||||
|
||||
This issue will be re-entered into triage for further investigation."
|
||||
|
||||
_post_comment "$ISSUE_NUMBER" "$COMMENT_BODY"
|
||||
log "Posted verification comment to issue #${ISSUE_NUMBER}"
|
||||
|
||||
# Re-trigger triage by adding in-triage label
|
||||
IN_TRIAGE_ID=$(_label_id "in-triage" "#d93f0b")
|
||||
_add_label "$ISSUE_NUMBER" "$IN_TRIAGE_ID"
|
||||
log "Re-triggered triage for issue #${ISSUE_NUMBER} — bug still reproduces"
|
||||
fi
|
||||
|
||||
# Exit after verification
|
||||
log "Verification complete. Outcome: ${VERIFY_OUTCOME}"
|
||||
exit 0
|
||||
else
|
||||
log "Not all sub-issues are closed yet — skipping verification"
|
||||
fi
|
||||
else
|
||||
log "Issue #${ISSUE_NUMBER} is not a parent with tracked sub-issues — running standard reproduction"
|
||||
fi
|
||||
fi
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# Read findings
|
||||
# ---------------------------------------------------------------------------
|
||||
|
|
@ -601,54 +920,6 @@ if find "$(dirname "${SCREENSHOT_PREFIX}")" -name "$(basename "${SCREENSHOT_PREF
|
|||
done
|
||||
fi
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# Label helpers
|
||||
# ---------------------------------------------------------------------------
|
||||
_label_id() {
|
||||
local name="$1" color="$2"
|
||||
local id
|
||||
id=$(curl -sf \
|
||||
-H "Authorization: token ${FORGE_TOKEN}" \
|
||||
"${FORGE_API}/labels" 2>/dev/null \
|
||||
| jq -r --arg n "$name" '.[] | select(.name == $n) | .id' 2>/dev/null || echo "")
|
||||
if [ -z "$id" ]; then
|
||||
id=$(curl -sf -X POST \
|
||||
-H "Authorization: token ${FORGE_TOKEN}" \
|
||||
-H "Content-Type: application/json" \
|
||||
"${FORGE_API}/labels" \
|
||||
-d "{\"name\":\"${name}\",\"color\":\"${color}\"}" 2>/dev/null \
|
||||
| jq -r '.id // empty' 2>/dev/null || echo "")
|
||||
fi
|
||||
echo "$id"
|
||||
}
|
||||
|
||||
_add_label() {
|
||||
local issue="$1" label_id="$2"
|
||||
[ -z "$label_id" ] && return 0
|
||||
curl -sf -X POST \
|
||||
-H "Authorization: token ${FORGE_TOKEN}" \
|
||||
-H "Content-Type: application/json" \
|
||||
"${FORGE_API}/issues/${issue}/labels" \
|
||||
-d "{\"labels\":[${label_id}]}" >/dev/null 2>&1 || true
|
||||
}
|
||||
|
||||
_remove_label() {
|
||||
local issue="$1" label_id="$2"
|
||||
[ -z "$label_id" ] && return 0
|
||||
curl -sf -X DELETE \
|
||||
-H "Authorization: token ${FORGE_TOKEN}" \
|
||||
"${FORGE_API}/issues/${issue}/labels/${label_id}" >/dev/null 2>&1 || true
|
||||
}
|
||||
|
||||
_post_comment() {
|
||||
local issue="$1" body="$2"
|
||||
curl -sf -X POST \
|
||||
-H "Authorization: token ${FORGE_TOKEN}" \
|
||||
-H "Content-Type: application/json" \
|
||||
"${FORGE_API}/issues/${issue}/comments" \
|
||||
-d "$(jq -nc --arg b "$body" '{body:$b}')" >/dev/null 2>&1 || true
|
||||
}
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# Apply labels and post findings
|
||||
# ---------------------------------------------------------------------------
|
||||
|
|
|
|||
|
|
@ -11,13 +11,19 @@
|
|||
#
|
||||
# timeout_minutes: hard upper bound on the Claude session.
|
||||
#
|
||||
# Exit gate logic:
|
||||
# Exit gate logic (standard mode):
|
||||
# 1. Can I reproduce it? → NO → rejected/blocked → EXIT
|
||||
# → YES → continue
|
||||
# 2. Is the cause obvious? → YES → in-progress + backlog issue → EXIT
|
||||
# → NO → in-triage → EXIT
|
||||
#
|
||||
# Turn budget: 60% on step 1 (reproduction), 40% on step 2 (cause check).
|
||||
# Exit gate logic (verification mode):
|
||||
# Triggered when all sub-issues of a parent bug-report are closed.
|
||||
# 1. Bug fixed → comment "verified fixed", remove in-progress, close issue
|
||||
# 2. Bug persists → comment "still reproduces", add in-triage, re-enter triage
|
||||
#
|
||||
# Turn budget (standard mode): 60% on step 1 (reproduction), 40% on step 2 (cause check).
|
||||
# Turn budget (verification mode): 100% on re-running reproduction steps.
|
||||
|
||||
name = "reproduce"
|
||||
description = "Primary: reproduce the bug. Secondary: check if cause is obvious. Exit gates enforced."
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue