Syncs test runs / results / cases between an automated test suite and TestRail (Gurock / Idera) - opens a Test Run for the build (`add_run`), batches per-case results back via `add_results_for_cases` (preferred over per-test `add_result_for_case` - N+1 API calls vs 1), maps the test framework's pass/fail/skip to TestRail status IDs, and attaches build URL + version + elapsed time. Use when the team's test management is standalone TestRail (not a Jira app) and automated suites must update it without a human copy-paste step; when the TCM is instead a Jira app use xray-integration (Xray) or zephyr-integration (Zephyr Scale), and for hosted cross-run flakiness analytics rather than TCM sync use currents-integration.
79
99%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
View guide
Low
Low-risk findings worth noting
Teams that use TestRail as the source of truth for test cases and runs need automation result sync - without it, automated runs don't update TestRail and the test-management view drifts from reality. This skill wires that sync.
The patterns below are the stable, long-documented TestRail API shapes; the canonical KB + API-reference URLs and the per-language client libraries are in References.
TestRail uses HTTP Basic auth with email + API key (preferred over password - the API key is per-user, revocable):
# Generated in TestRail: My Settings → API Keys
TESTRAIL_API_KEY=<generated>
TESTRAIL_USER=test-runner@example.com
TESTRAIL_HOST=https://yourcompany.testrail.ioAll requests use:
Authorization: Basic <base64(email:api_key)>
Content-Type: application/jsonThe base API URL is ${TESTRAIL_HOST}/index.php?/api/v2. Every
endpoint is appended after ?/api/v2.
Two common patterns:
def test_C1234_can_add_to_cart():
...A regex extracts C1234 (the TestRail case ID) at sync time.
@Test
@TestRailCase(id = 1234)
void canAddToCart() { ... }Or in JS:
test('can add to cart [C1234]', async () => {
// ...
});Pattern A is the lowest-friction; Pattern B is cleaner when the test framework supports custom annotations. Either way, the sync script needs a way to find the case ID from the test result.
# scripts/testrail_sync.py
import base64, json, requests
from os import environ as env
API = f"{env['TESTRAIL_HOST']}/index.php?/api/v2"
AUTH = base64.b64encode(f"{env['TESTRAIL_USER']}:{env['TESTRAIL_API_KEY']}".encode()).decode()
HEADERS = {'Authorization': f'Basic {AUTH}', 'Content-Type': 'application/json'}
def open_run(project_id, suite_id, name, case_ids):
r = requests.post(
f'{API}/add_run/{project_id}',
headers=HEADERS,
json={
'suite_id': suite_id,
'name': name,
'include_all': False,
'case_ids': case_ids,
},
)
r.raise_for_status()
return r.json()['id'] # Run IDinclude_all: False + case_ids: [...] opens a run scoped to the
exact cases the automated suite covers. Without this, a 5,000-case
project produces a 5,000-row Test Run with thousands of empty
cells.
The well-known status ID convention for stock TestRail installations:
| Status | ID |
|---|---|
| Passed | 1 |
| Blocked | 2 |
| Untested | 3 |
| Retest | 4 |
| Failed | 5 |
Custom status IDs (added by the project admin) follow 6+. Read the
get_statuses endpoint at sync-script init to confirm - don't
hard-code.
Verify before batching: call get_statuses and assert every
status_id in your map appears in the returned set. If one is missing
(a renamed or custom status), fix the map and re-run rather than
posting - TestRail accepts an unknown status_id and writes the result
to the wrong status silently.
def add_results(run_id, results):
"""results = [{'case_id': 1234, 'status_id': 1, 'comment': '...', 'elapsed': '12s'}]"""
r = requests.post(
f'{API}/add_results_for_cases/{run_id}',
headers=HEADERS,
json={'results': results},
)
r.raise_for_status()
return r.json()Use add_results_for_cases (batch), not add_result_for_case
(per-case). A 200-test run is one POST instead of 200 POSTs;
TestRail's rate limit (180 req/min on shared cloud) makes per-case
posting flaky.
Each result entry accepts case_id + status_id (both required),
plus optional comment, elapsed, version, defects, and
assignedto_id. The full per-result field list is in
references/ci-and-fields.md.
After all results are in:
def close_run(run_id):
requests.post(f'{API}/close_run/{run_id}', headers=HEADERS)Closed runs are read-only - no further results can be added. Useful for release-stamp runs; skip for runs that get re-run.
Run the sync as an if: always() step after the test step so failed
runs still update TestRail. The script parses junit.xml
(junit-xml-analysis), extracts case IDs (Step 2), opens a run scoped
to the covered cases (Step 3), batches results (Step 4), and optionally
closes the run on main only (Step 5). Supply TESTRAIL_HOST /
TESTRAIL_USER / TESTRAIL_API_KEY from CI secrets. The full GitHub
Actions workflow, the per-result field list, and handling for tests that
carry no case ID are in
references/ci-and-fields.md.
A Jest suite syncing one build to TestRail project 42, suite 7:
test('can add to cart [C1234]', ...).export TESTRAIL_HOST=https://acme.testrail.io
export TESTRAIL_USER=ci@acme.com
export TESTRAIL_API_KEY=... # My Settings > API Keys
npm test -- --reporters=jest-junitrun_id = open_run(42, 7, "main · a1b2c3d", [1234]) # returns the Run ID
add_results(run_id, [
{'case_id': 1234, 'status_id': 1, 'comment': 'green on CI', 'elapsed': '12s'},
])Open the run in TestRail to see C1234 marked Passed with the 12s
elapsed time. Close the run (Step 5) only when this is the main build.
| Anti-pattern | Why it fails | Fix |
|---|---|---|
Per-test add_result_for_case calls | N API calls; rate limit (180 req/min on Cloud) trips on suites >180 cases. | add_results_for_cases batch (Step 4). |
Hard-coded status IDs without get_statuses | Custom statuses break the mapping; "Failed" goes to "Custom Status" silently. | Fetch get_statuses at script init; build the map dynamically. |
include_all: True on add_run | The run includes every case in the suite, most as Untested; runs become noise. | include_all: False + explicit case_ids: [...]. |
| Posting credentials as URL params | Secrets leak in proxy logs. | Always Basic auth header (Step 1). |
| No retry on 5xx | TestRail Cloud has occasional 502s; one transient failure loses the whole run. | Retry with exponential backoff on 5xx; cap at 3 attempts. |
| Closing every run, including PR runs | Closed runs can't accept reruns; a PR retest after fixing flake fails to update. | Close only main runs (Step 6); PR runs stay open. |
| Storing case IDs in test code AND in TestRail | Two sources of truth; renames drift. | TestRail is canonical; test code references via ID only (Step 2 Pattern A). |
add_case before the first sync; the sync script
doesn't create cases on the fly.custom_step_results
is per-installation; the project admin must enable a custom field
for steps before the API accepts step-level data.https://support.testrail.com/hc/en-us/categories/7080117421716
(categorized KB; per-endpoint articles).https://support.testrail.com/hc/en-us/sections/7077986539540
(Results / Runs / Cases / Statuses endpoints).testrail (Python),
testrail-java-client (Java), testrail-api (JS) - reference
implementations of the API shapes above.junit-xml-analysis -
upstream parser for the input the sync script consumes.xray-integration,
zephyr-integration - sibling
Jira-native alternatives.currents-integration -
different role: test analytics over time, not test management.