RomanNum3ral
/
agentic-os
mirror of https://github.com/modimihir07/agentic-os.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Merge pull request #12 from zumayaaustin-creator/devin/1783552096-error-handling 

* Improve error handling: propagate corrupt-JSON errors, stop swallowing failures

Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>

* Validate skill name to prevent path traversal (CodeQL)

Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>

* Tighten skill name allowlist to exclude '.' (path traversal)

Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>

* Resolve skill name via directory match to break path-injection taint (CodeQL)

Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>

* Make aggregate listings tolerate a corrupt file (best_effort), keep single GETs strict

Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>

* Move kanban_task_path inside try block in daemon thread

* Resolve existing skills via iterdir match to break path-injection taint (CodeQL)

Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>

* Recompute path in kanban dispatch error handler to avoid unbound local

Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>

* Fix unbound variable in _run_kanban_agent error handler

* Restore kanban dispatch success branch mangled by misapplied review suggestion

Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>

---------

Co-authored-by: zumayaaustin <zumayaaustin@gmail.com>
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
This commit is contained in:
zumayaaustin-creator 2026-07-08 18:42:15 -07:00 committed by GitHub
parent 2d80ab7348 4604b433e8
commit 1a2e0b2902
2 changed files with 151 additions and 77 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

40
scheduler/scheduler.py
View File

@ -24,24 +24,40 @@ def run_skill(skill_name: str):
"skill": skill_name,
"timestamp": datetime.now(timezone.utc).isoformat(),
}
with open(audit_file, "a") as f:
f.write(json.dumps(entry) + "\n")
try:
audit_file.parent.mkdir(parents=True, exist_ok=True)
with open(audit_file, "a", encoding="utf-8") as f:
f.write(json.dumps(entry) + "\n")
except OSError as e:
print(f" [audit] failed to record run of {skill_name!r}: {e}")
print(f"[{datetime.now().isoformat()}] Ran skill: {skill_name}")
def load_jobs(scheduler: BackgroundScheduler):
"""Load job definitions from jobs/ directory."""
"""Load job definitions from jobs/ directory.
A single malformed job file is logged and skipped rather than being allowed
to abort loading of every other job.
"""
for job_file in JOBS_DIR.glob("*.json"):
data = json.loads(job_file.read_text())
try:
data = json.loads(job_file.read_text())
except (json.JSONDecodeError, OSError) as e:
print(f" Skipping {job_file.name}: could not read job ({e})")
continue
if not data.get("enabled", True):
continue
scheduler.add_job(
run_skill,
CronTrigger.from_crontab(data["cron"]),
args=[data["skill"]],
id=data.get("id", data["name"]),
name=data["name"],
replace_existing=True,
)
try:
scheduler.add_job(
run_skill,
CronTrigger.from_crontab(data["cron"]),
args=[data["skill"]],
id=data.get("id", data["name"]),
name=data["name"],
replace_existing=True,
)
except (KeyError, ValueError) as e:
print(f" Skipping {job_file.name}: invalid job definition ({e})")
continue
print(f" Scheduled: {data['name']} ({data['cron']})")
def main():

188
server.py
View File

@ -133,11 +133,24 @@ def list_dir(path: Path):
return []
return sorted([p.name for p in path.iterdir() if not p.name.startswith(".")])
def read_json(path: Path, default=None):
"""Load JSON from path, returning default when the file is missing."""
def read_json(path: Path, default=None, best_effort=False):
"""Load JSON from path, returning ``default`` when the file is missing.
On corrupt or unreadable content a descriptive ``HTTPException(500)`` is
raised so the error is propagated to the client instead of surfacing as an
opaque 500. Aggregate/listing callers can set ``best_effort=True`` to
tolerate one bad file: the corruption is logged and ``default`` is returned
instead of aborting the whole view.
"""
if not path.exists():
return default
return json.loads(path.read_text(encoding="utf-8"))
try:
return json.loads(path.read_text(encoding="utf-8"))
except (json.JSONDecodeError, OSError, UnicodeDecodeError) as e:
if best_effort:
print(f"[load] skipping corrupt {path.name}: {e}")
return default
raise HTTPException(500, f"Failed to read {path.name}: {e}")
def write_json(path: Path, data, indent: int = 2):
"""Serialize data as pretty JSON to path."""
@ -162,8 +175,14 @@ def append_audit(entry: dict):
audit_file = BASE_DIR / "audit" / "audit.log"
entry["timestamp"] = get_timestamp()
entry["id"] = new_id()
with open(audit_file, "a") as f:
f.write(json.dumps(entry) + "\n")
try:
audit_file.parent.mkdir(parents=True, exist_ok=True)
with open(audit_file, "a", encoding="utf-8") as f:
f.write(json.dumps(entry) + "\n")
except OSError as e:
# Auditing is best-effort: never let a logging failure abort the
# underlying operation, but surface it on the server console.
print(f"[audit] failed to write entry {entry.get('action')!r}: {e}")
# ─── Agent Discovery (instant filesystem checks) ────────────────────
@ -200,22 +219,24 @@ def hermes_available() -> bool:
def check_agent(name: str) -> dict:
"""Filesystem-based check for opencode/gemini; hermes needs a real subprocess since it may live inside WSL."""
try:
if name == "opencode":
exists = shutil.which("opencode") is not None
status = "online" if exists else "offline"
elif name == "hermes":
exists = hermes_available()
status = "online" if exists else "offline"
elif name == "gemini":
# Gemini has valid OAuth tokens logged in
oauth = Path.home() / ".gemini" / "oauth_creds.json"
exists = shutil.which("gemini") is not None
logged_in = oauth.exists() and "ya29" in oauth.read_text()
status = "online" if exists and logged_in else "offline" if not exists else "warning"
else:
status = "offline"
except Exception:
if name == "opencode":
status = "online" if shutil.which("opencode") is not None else "offline"
elif name == "hermes":
status = "online" if hermes_available() else "offline"
elif name == "gemini":
exists = shutil.which("gemini") is not None
# Gemini needs a valid OAuth token on disk to be usable.
logged_in = False
oauth = Path.home() / ".gemini" / "oauth_creds.json"
if oauth.exists():
try:
logged_in = "ya29" in oauth.read_text()
except OSError as e:
# Distinguish an unreadable credential file from "not logged in"
# instead of silently reporting the agent as offline.
print(f"[agent-health] could not read gemini credentials: {e}")
status = "online" if exists and logged_in else "offline" if not exists else "warning"
else:
status = "offline"
return {"name": name, "status": status}
@ -272,6 +293,21 @@ def skill_dir_path(name: str) -> Path:
raise HTTPException(400, "Invalid skill name")
return candidate
def resolve_skill_dir(name: str) -> Path:
"""Return the directory of an existing skill by matching ``name`` against the
actual directory entries.
Using the entry from ``iterdir()`` (rather than a path built from ``name``)
means traversal input can never escape the skills directory. Raises 404 if no
skill matches.
"""
base = BASE_DIR / "skills"
if base.exists():
for entry in base.iterdir():
if entry.is_dir() and entry.name == name:
return entry
raise HTTPException(404, "Skill not found")
def skill_context_file_path(name: str, filename: str) -> Path:
if not SKILL_CONTEXT_FILENAME_RE.fullmatch(filename or ""):
raise HTTPException(400, "Invalid file name")
@ -287,8 +323,8 @@ def list_skills():
for d in iter_skill_dirs():
skill_md = read_file(d / "SKILL.md")
learnings = read_file(d / "learnings.md")
eval_data = read_json(d / "eval.json", {})
score_history = read_json(d / "score-history.json", [])
eval_data = read_json(d / "eval.json", {}, best_effort=True)
score_history = read_json(d / "score-history.json", [], best_effort=True)
skills.append({
"name": d.name,
"description": skill_md[:200] if skill_md else "",
@ -300,15 +336,13 @@ def list_skills():
@app.get("/api/skills/{name}")
def get_skill(name: str):
path = skill_dir_path(name)
if not path.exists():
raise HTTPException(404, "Skill not found")
path = resolve_skill_dir(name)
return {
"name": name,
"skill": read_file(path / "SKILL.md"),
"learnings": read_file(path / "learnings.md"),
"eval": json.loads((path / "eval.json").read_text()) if (path / "eval.json").exists() else {},
"score_history": json.loads((path / "score-history.json").read_text()) if (path / "score-history.json").exists() else [],
"eval": read_json(path / "eval.json", default={}),
"score_history": read_json(path / "score-history.json", default=[]),
"context": [f.name for f in (path / "context").iterdir()] if (path / "context").exists() else [],
}
@ -360,9 +394,7 @@ def delete_skill_context_file(name: str, filename: str):
@app.post("/api/skills/{name}/run")
def run_skill(name: str, req: Optional[SkillRunRequest] = None):
path = skill_dir_path(name)
if not path.exists():
raise HTTPException(404, "Skill not found")
path = resolve_skill_dir(name)
agent_choice = req.agent if req else "auto"
skill_input = req.input if req else ""
@ -443,10 +475,8 @@ def run_skill(name: str, req: Optional[SkillRunRequest] = None):
@app.get("/api/skills/{name}/eval")
def get_skill_eval(name: str):
path = skill_dir_path(name) / "score-history.json"
if not path.exists():
return {"scores": []}
return {"scores": json.loads(path.read_text())}
path = resolve_skill_dir(name) / "score-history.json"
return {"scores": read_json(path, default=[])}
# ─── Routes: Scheduler ────────────────────────────────────────────
@ -455,7 +485,9 @@ def list_jobs():
jobs_dir = BASE_DIR / "scheduler" / "jobs"
jobs = []
for f in sorted(jobs_dir.glob("*.json")):
jobs.append(json.loads(f.read_text()))
job = read_json(f, default=None, best_effort=True)
if job is not None:
jobs.append(job)
return jobs
@app.post("/api/scheduler/jobs")
@ -482,8 +514,8 @@ def create_job(job: ScheduleJobRequest):
def delete_job(job_id: str):
jobs_dir = BASE_DIR / "scheduler" / "jobs"
for f in jobs_dir.glob("*.json"):
data = json.loads(f.read_text())
if data.get("id") == job_id:
data = read_json(f, default=None, best_effort=True)
if data and data.get("id") == job_id:
f.unlink()
append_audit({"action": "job_deleted", "job_id": job_id})
return {"status": "deleted"}
@ -497,7 +529,15 @@ def get_audit(limit: int = Query(100, le=500)):
if not audit_file.exists():
return {"entries": []}
lines = audit_file.read_text().strip().split("\n")
entries = [json.loads(l) for l in lines if l.strip()]
entries = []
for l in lines:
if not l.strip():
continue
try:
entries.append(json.loads(l))
except json.JSONDecodeError:
# Skip a corrupt line rather than failing the whole audit view.
continue
return {"entries": entries[-limit:]}
# ─── Routes: Cost Analytics ───────────────────────────────────────
@ -996,7 +1036,9 @@ def load_kanban_tasks():
ensure_dir(KANBAN_DIR)
tasks = []
for f in sorted(KANBAN_DIR.glob("*.json")):
tasks.append(json.loads(f.read_text()))
task = read_json(f, default=None, best_effort=True)
if task is not None:
tasks.append(task)
return tasks
KANBAN_ID_RE = re.compile(r"^[0-9a-f]{6,16}$")
@ -1034,33 +1076,49 @@ def dispatch_kanban_task(task_id: str):
threading.Thread(target=_run_kanban_agent, args=(task_id,), daemon=True).start()
def _run_kanban_agent(task_id: str):
path = kanban_task_path(task_id)
if not path.exists():
return
task = json.loads(path.read_text())
agent = task.get("assignee")
prompt = task["title"] if not task.get("body") else f"{task['title']}\n\n{task['body']}"
# Runs in a daemon thread: any unhandled exception would be lost and leave
# the task stuck in "in_progress" forever, so catch failures and surface
# them by marking the task blocked with the error.
try:
path = kanban_task_path(task_id)
if not path.exists():
return
task = json.loads(path.read_text())
agent = task.get("assignee")
prompt = task["title"] if not task.get("body") else f"{task['title']}\n\n{task['body']}"
response = execute_agent(agent, prompt)
failed = response.startswith(("⏱", "⚠", "Unknown agent"))
response = execute_agent(agent, prompt)
failed = response.startswith(("⏱", "⚠", "Unknown agent"))
task = json.loads(path.read_text()) # reload in case it changed while the agent ran
task.setdefault("comments", []).append({
"id": new_id(),
"message": f"🤖 **{agent}**\n\n{response}",
"timestamp": get_timestamp(),
})
if failed:
task["status"] = "blocked"
task["block_reason"] = response[:300]
append_audit({"action": "kanban_task_dispatch_failed", "task_id": task_id, "agent": agent})
else:
task["status"] = "done"
task["summary"] = response[:300]
task["completed_at"] = get_timestamp()
append_audit({"action": "kanban_task_dispatch_completed", "task_id": task_id, "agent": agent})
task["updated"] = get_timestamp()
save_kanban_task(task)
task = json.loads(path.read_text()) # reload in case it changed while the agent ran
task.setdefault("comments", []).append({
"id": new_id(),
"message": f"🤖 **{agent}**\n\n{response}",
"timestamp": get_timestamp(),
})
if failed:
task["status"] = "blocked"
task["block_reason"] = response[:300]
append_audit({"action": "kanban_task_dispatch_failed", "task_id": task_id, "agent": agent})
else:
task["status"] = "done"
task["summary"] = response[:300]
task["completed_at"] = get_timestamp()
append_audit({"action": "kanban_task_dispatch_completed", "task_id": task_id, "agent": agent})
task["updated"] = get_timestamp()
save_kanban_task(task)
except Exception as e:
print(f"[kanban] dispatch for task {task_id} crashed: {e}")
try:
path = kanban_task_path(task_id)
task = json.loads(path.read_text())
task["status"] = "blocked"
task["block_reason"] = f"Dispatch crashed: {e}"[:300]
task["updated"] = get_timestamp()
save_kanban_task(task)
append_audit({"action": "kanban_task_dispatch_error", "task_id": task_id, "error": str(e)[:200]})
except Exception as inner:
print(f"[kanban] could not mark task {task_id} as blocked: {inner}")
def load_goals():
return read_json(GOALS_FILE, [])