K-Dense-AI/scientific-agent-skills / skills/iso-standards-readiness
iso-standards-readiness
VerifiedPrepares and structurally reviews readiness evidence for ISO management-system and laboratory-competence standards - ISO 13485 medical device QMS, ISO 14971 device risk management, ISO/IEC 17025 testing and calibration laboratories, and ISO 15189 medical laboratories. Use when organizing declared scope, controlled documents, risk-management files, scope of accreditation, traceability, CAPA, external-provider controls, or bounded local evidence manifests, and when separating ISO certification from laboratory accreditation, FDA QMSR inspection, CLIA certification, MDSAP, and EU MDR/IVDR evidence boundaries. Not for legal applicability, compliance, certification, or accreditation decisions; contains no clause text.
scripts/check_capa.py
View on GitHub#!/usr/bin/env python3
"""Check CAPA record structure, evidence, and effectiveness gates."""
from __future__ import annotations
from typing import Any
from _common import (
Review,
finish,
guarded_main,
load_json,
require_root_object,
standard_parser,
)
CAPA_STATUSES = {
"open",
"investigating",
"actions-in-progress",
"effectiveness-pending",
"closed",
"cancelled",
}
EFFECTIVENESS_RESULTS = {"pending", "effective", "ineffective"}
RISK_DECISIONS = {"escalate", "control-and-monitor", "no-escalation"}
def _text_array(
review: Review,
value: Any,
path: str,
*,
min_items: int = 1,
) -> list[str]:
raw = review.list(value, path, min_items=min_items)
if raw is None:
return []
result: list[str] = []
for index, item in enumerate(raw):
if not isinstance(item, str) or not item.strip():
review.add(
"TEXT_REQUIRED",
f"{path}[{index}]",
"requires non-placeholder text",
)
else:
result.append(item.strip())
return result
def validate(data: dict[str, Any]) -> tuple[Review, dict[str, int]]:
review = Review()
metadata = review.object(data.get("metadata"), "metadata")
if metadata is not None:
review.controlled_item(metadata, "metadata", require_approved=True)
review.text(metadata, "register_id", "metadata", max_chars=120)
review.date(metadata, "review_date", "metadata")
capas = review.list(data.get("capas"), "capas", min_items=1) or []
closed = 0
for index, value in enumerate(capas):
path = f"capas[{index}]"
capa = review.object(value, path)
if capa is None:
continue
review.text(capa, "id", path, max_chars=120)
review.text(capa, "owner", path, max_chars=200)
status = review.choice(capa, "status", CAPA_STATUSES, path)
review.text(capa, "source_event", path, max_chars=1_000)
review.text(capa, "problem_statement", path, max_chars=2_000)
review.text(capa, "scope", path, max_chars=2_000)
review.text(capa, "correction_or_containment", path, max_chars=2_000)
review.evidence(capa, path)
review.source_refs(capa, path)
review.approval(capa, path, require_approved=status in {"closed", "cancelled"})
risk_path = f"{path}.risk_assessment"
risk = review.object(capa.get("risk_assessment"), risk_path)
if risk is not None:
review.text(risk, "patient_and_user_impact", risk_path, max_chars=2_000)
review.text(risk, "product_and_process_impact", risk_path, max_chars=2_000)
review.text(risk, "reportability_review", risk_path, max_chars=2_000)
review.choice(risk, "decision", RISK_DECISIONS, risk_path)
review.text(risk, "decision_rationale", risk_path, max_chars=2_000)
review.text(risk, "owner", risk_path, max_chars=200)
review.evidence(risk, risk_path)
review.approval(risk, risk_path, require_approved=True)
investigation_path = f"{path}.investigation"
investigation = review.object(capa.get("investigation"), investigation_path)
if investigation is not None:
review.text(investigation, "method", investigation_path, max_chars=500)
review.text(
investigation,
"root_cause_or_justified_conclusion",
investigation_path,
max_chars=3_000,
)
review.text(
investigation,
"systemic_extent_review",
investigation_path,
max_chars=2_000,
)
review.text(investigation, "owner", investigation_path, max_chars=200)
review.evidence(investigation, investigation_path)
review.approval(
investigation,
investigation_path,
require_approved=status not in {"open", "investigating"},
)
actions = review.list(capa.get("actions"), f"{path}.actions", min_items=1) or []
all_implemented = True
for action_index, action_value in enumerate(actions):
action_path = f"{path}.actions[{action_index}]"
action = review.object(action_value, action_path)
if action is None:
all_implemented = False
continue
review.text(action, "id", action_path, max_chars=120)
review.text(action, "description", action_path, max_chars=2_000)
review.text(action, "owner", action_path, max_chars=200)
review.date(action, "due_date", action_path)
implemented = action.get("implemented_date")
if implemented in (None, ""):
all_implemented = False
if status in {"effectiveness-pending", "closed"}:
review.add(
"ACTION_NOT_IMPLEMENTED",
f"{action_path}.implemented_date",
"implemented date is required at this CAPA status",
)
else:
review.date(action, "implemented_date", action_path)
review.evidence(action, action_path)
review.approval(action, action_path, require_approved=True)
review.unique_ids(actions, f"{path}.actions")
_text_array(
review,
capa.get("change_control_refs"),
f"{path}.change_control_refs",
min_items=1,
)
effect_path = f"{path}.effectiveness"
effectiveness = review.object(capa.get("effectiveness"), effect_path)
result = None
if effectiveness is not None:
review.text(effectiveness, "plan", effect_path, max_chars=2_000)
review.text(
effectiveness,
"objective_acceptance_criteria",
effect_path,
max_chars=2_000,
)
review.text(effectiveness, "baseline", effect_path, max_chars=1_000)
review.text(
effectiveness,
"sample_or_observation_window",
effect_path,
max_chars=1_000,
)
review.date(effectiveness, "due_date", effect_path)
review.text(effectiveness, "owner", effect_path, max_chars=200)
review.text(
effectiveness,
"independent_reviewer",
effect_path,
max_chars=200,
)
result = review.choice(
effectiveness,
"result",
EFFECTIVENESS_RESULTS,
effect_path,
)
if result != "pending":
review.text(
effectiveness,
"conclusion",
effect_path,
max_chars=2_000,
)
review.date(effectiveness, "review_date", effect_path)
review.evidence(effectiveness, effect_path)
review.approval(
effectiveness,
effect_path,
require_approved=True,
)
if status == "closed":
closed += 1
if not all_implemented:
review.add(
"CLOSURE_BLOCKED",
path,
"closed CAPA has unimplemented actions",
"blocker",
)
if result != "effective":
review.add(
"CLOSURE_BLOCKED",
f"{effect_path}.result",
"closed CAPA requires approved effective result",
"blocker",
)
review.date(capa, "closure_date", path)
review.text(capa, "closure_summary", path, max_chars=2_000)
elif result == "ineffective":
review.add(
"REOPEN_REQUIRED",
f"{effect_path}.result",
"ineffective action must not be closed and needs documented follow-up",
"blocker",
)
if status == "cancelled":
review.text(capa, "cancellation_rationale", path, max_chars=2_000)
review.unique_ids(capas, "capas")
return review, {"capas": len(capas), "closed_capas": closed}
def main() -> int:
parser = standard_parser(
"Check CAPA records and fail closed on missing effectiveness evidence.",
"Path to the local CAPA JSON file",
)
args = parser.parse_args()
data = require_root_object(load_json(args.input))
review, metrics = validate(data)
return finish("check_capa", review, args, metrics=metrics)
if __name__ == "__main__":
guarded_main(main)
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| finishing-a-development-branchobra/superpowersUse when implementation is complete, all tests pass, and you need to decide how to integrate the work | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| systematic-debuggingobra/superpowersUse when encountering any bug, test failure, or unexpected behavior, before proposing fixes | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| test-driven-developmentobra/superpowersUse when implementing any feature or bugfix, before writing implementation code | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| skill-creatoranthropics/skillsCreate new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |