# Send Yandex Tracker to your agent Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually. ## Fast path - Download the package from Yavira. - Extract it into a folder your agent can access. - Paste one of the prompts below and point your agent at the extracted folder. ## Suggested prompts ### New install ```text I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Then review README.md for any prerequisites, environment setup, or post-install checks. Tell me what you changed and call out any manual steps you could not complete. ``` ### Upgrade existing ```text I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Then review README.md for any prerequisites, environment setup, or post-install checks. Summarize what changed and any follow-up checks I should run. ``` ## Machine-readable fields ```json { "schemaVersion": "1.0", "item": { "slug": "yandex-tracker", "name": "Yandex Tracker", "source": "tencent", "type": "skill", "category": "开发工具", "sourceUrl": "https://clawhub.ai/Kandler3/yandex-tracker", "canonicalUrl": "https://clawhub.ai/Kandler3/yandex-tracker", "targetPlatform": "OpenClaw" }, "install": { "downloadUrl": "/downloads/yandex-tracker", "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=yandex-tracker", "sourcePlatform": "tencent", "targetPlatform": "OpenClaw", "packageFormat": "ZIP package", "primaryDoc": "SKILL.md", "includedAssets": [ "README.md", "SKILL.md" ], "downloadMode": "redirect", "sourceHealth": { "source": "tencent", "status": "healthy", "reason": "direct_download_ok", "recommendedAction": "download", "checkedAt": "2026-05-07T17:22:31.273Z", "expiresAt": "2026-05-14T17:22:31.273Z", "httpStatus": 200, "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=afrexai-annual-report", "contentType": "application/zip", "probeMethod": "head", "details": { "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=afrexai-annual-report", "contentDisposition": "attachment; filename=\"afrexai-annual-report-1.0.0.zip\"", "redirectLocation": null, "bodySnippet": null }, "scope": "source", "summary": "Source download looks usable.", "detail": "Yavira can redirect you to the upstream package for this source.", "primaryActionLabel": "Download for OpenClaw", "primaryActionHref": "/downloads/yandex-tracker" }, "validation": { "installChecklist": [ "Use the Yavira download entry.", "Review SKILL.md after the package is downloaded.", "Confirm the extracted package contains the expected setup assets." ], "postInstallChecks": [ "Confirm the extracted package includes the expected docs or setup files.", "Validate the skill or prompts are available in your target agent workspace.", "Capture any manual follow-up steps the agent could not complete." ] } }, "links": { "detailUrl": "https://openagent3.xyz/skills/yandex-tracker", "downloadUrl": "https://openagent3.xyz/downloads/yandex-tracker", "agentUrl": "https://openagent3.xyz/skills/yandex-tracker/agent", "manifestUrl": "https://openagent3.xyz/skills/yandex-tracker/agent.json", "briefUrl": "https://openagent3.xyz/skills/yandex-tracker/agent.md" } } ``` ## Documentation ### Yandex Tracker (Python Client) Use yandex_tracker_client to interact with Yandex Tracker API v2. ### How to work with this skill Write and execute Python scripts to fulfill user requests. The workflow: Write a self-contained Python script that initializes the client, performs all needed API calls, aggregates and formats the result, then prints it. Save to /tmp/tracker_script.py and run: python3 /tmp/tracker_script.py For simple one-liners it's fine to use python3 -c "...", but prefer a file for anything multi-step. Aggregation: The API returns lazy iterables — always collect into lists when you need to count, sort, filter, or display summaries. Combine multiple queries in one script (e.g. fetch issues then fetch comments for each, or join data from two queues) rather than making separate tool calls. Print structured output so the result is easy to read. # Example: aggregate issues by assignee across a queue issues = list(client.issues.find(filter={'queue': 'QUEUE'}, per_page=100)) from collections import Counter counts = Counter(str(i.assignee) for i in issues) for assignee, n in counts.most_common(): print(f'{n:3d} {assignee}') ### Credentials Required env (declare in skill metadata; set in openclaw.json → env): TRACKER_TOKEN — Required. Use a least-privilege OAuth token (oauth.yandex.ru) with only Tracker scope, or a temporary IAM token for Yandex Cloud. Do not use broad admin tokens. One of: TRACKER_ORG_ID (Yandex 360, numeric) or TRACKER_CLOUD_ORG_ID (Yandex Cloud, string) ### Client initialization (boilerplate — always include) import os from yandex_tracker_client import TrackerClient token = os.environ['TRACKER_TOKEN'] org_id = os.environ.get('TRACKER_ORG_ID') cloud_org_id = os.environ.get('TRACKER_CLOUD_ORG_ID') if cloud_org_id: client = TrackerClient(token=token, cloud_org_id=cloud_org_id) else: client = TrackerClient(token=token, org_id=int(org_id)) ### Get issue by key issue = client.issues['QUEUE-42'] print(issue.key, issue.summary, issue.status.id, issue.assignee.login if issue.assignee else None) ### Custom fields Real queues almost always have custom fields (story points, business fields, etc.). Their keys look like customFieldId (camelCase) and are queue-specific. # Read — access by attribute name (same as standard fields) print(issue.storyPoints) # returns None if absent print(issue.as_dict()) # dump all fields including custom ones — use to discover keys # Update issue.update(storyPoints=5, myCustomField='value') # Filter by custom field in find() issues = client.issues.find(filter={'queue': 'QUEUE', 'storyPoints': {'from': 3}}) # Discover all fields available in a queue (id is the key to use) for f in client.fields.get_all(): print(f.id, f.name) Use issue.as_dict() on a real issue to discover which custom field keys the queue uses before writing update/filter code. ### issues.find() — search # Tracker Query Language string (copy from Tracker UI) issues = client.issues.find('Queue: QUEUE Assignee: me() Status: inProgress') # Structured filter (dict) issues = client.issues.find( filter={ 'queue': 'QUEUE', # queue key 'assignee': 'user_login', # login or 'me()' 'author': 'user_login', 'status': 'inProgress', # status .id 'type': 'bug', # issue type .id 'priority': 'critical', # priority .id 'tags': ['backend', 'urgent'], # all tags must match 'created': {'from': '2026-01-01', 'to': '2026-02-01'}, 'updated': {'from': '2026-01-15'}, 'deadline': {'to': '2026-03-01'}, 'followers': 'user_login', 'components': 'component_name', }, order=['-updatedAt', '+priority'], # prefix: - desc, + asc per_page=100, # max 100 per page; pagination is automatic ) # Iterating auto-fetches all pages; wrap in list() to materialise issues = list(issues) # Batch fetch specific issues by key issues = list(client.issues.find(keys=['QUEUE-1', 'QUEUE-2', 'QUEUE-3'])) ### issues.create() issue = client.issues.create( queue='QUEUE', # required summary='Bug: login fails', # required type={'name': 'Bug'}, # or {'id': 'bug'} — use client.issue_types.get_all() description='Steps...', assignee='user_login', priority='critical', # id: 'blocker','critical','major','normal','minor','trivial' followers=['login1', 'login2'], tags=['backend', 'urgent'], components=['component_name'], parent='QUEUE-10', # parent issue key sprint={'id': 123}, # sprint id ) print(issue.key) ### issue.update() issue.update( summary='New title', description='Updated text', assignee='other_login', priority='minor', # Lists: pass full replacement OR mutation dict tags=['new_tag'], # replace entirely tags={'add': ['tag1'], 'remove': ['tag2']}, # partial mutation followers={'add': ['login1']}, components={'add': ['comp'], 'remove': []}, ) ### issue.transitions — status changes # Always list first — transition IDs are queue-specific, never guess for t in issue.transitions.get_all(): print(t.id, t.to.id, t.to.display) # Execute — all kwargs optional issue.transitions['close'].execute( comment='Fixed in v2.3', resolution='fixed', # 'fixed','wontFix','duplicate','invalid','later' — queue-dependent ) ### issue.comments for c in list(issue.comments.get_all()): print(c.id, c.createdBy.login, c.text) # Create issue.comments.create( text='Fixed in v2.3', summonees=['login1', 'login2'], # triggers @mention notification attachments=['path/to/file.png'], # file paths — auto-uploaded, converted to IDs ) issue.comments[42].update(text='Corrected note', summonees=['login1']) issue.comments[42].delete() ### issue.links for link in issue.links: print(link.type.id, link.direction, link.object.key) # relationship values (standard): # 'relates', 'blocks', 'is blocked by', # 'duplicates', 'is duplicated by', # 'depends on', 'is dependent of', # 'is subtask for', 'is parent task for' issue.links.create(issue='OTHER-10', relationship='relates') issue.links[42].delete() ### issue.attachments for a in issue.attachments: print(a.id, a.name, a.mimetype, a.size) issue.attachments.create('/path/to/file.txt') # upload by path a.download_to('/tmp/') # download to dir issue.attachments[42].delete() ### issue.worklog # Create worklog entry issue.worklog.create( duration='PT1H30M', # ISO 8601: PT30M, PT2H, P1D, P1DT2H30M comment='Fixed auth bug', # optional start='2026-02-24T10:00:00+03:00', # optional, defaults to now ) for w in list(issue.worklog.get_all()): print(w.id, w.duration, w.comment, w.createdBy.login) issue.worklog[42].update(duration='PT2H', comment='Revised estimate') issue.worklog[42].delete() # Fetch worklogs across multiple issues at once entries = client.worklog.find(issue=['QUEUE-1', 'QUEUE-2'], createdBy='me()') ### Queues queue = client.queues['QUEUE'] print(queue.key, queue.name, queue.lead.login) for q in client.queues.get_all(): print(q.key, q.name) ### Bulk operations # issues arg: list of keys OR list of issue objects from find() # Bulk update — any issue field as kwarg bc = client.bulkchange.update( ['QUEUE-1', 'QUEUE-2', 'QUEUE-3'], priority='minor', assignee='user_login', tags={'add': ['reviewed'], 'remove': ['draft']}, ) bc.wait() print(bc.status) # 'COMPLETE' or 'FAILED' # Bulk transition — transition id + optional field values bc = client.bulkchange.transition( ['QUEUE-1', 'QUEUE-2'], 'close', # transition id resolution='wontFix', # optional extra fields ) bc.wait() # Bulk move to another queue bc = client.bulkchange.move( ['QUEUE-1', 'QUEUE-2'], 'NEWQUEUE', move_all_fields=False, # copy all field values move_to_initial_status=False, # reset to initial status ) bc.wait() ### Object fields reference All objects are dynamic — accessing a missing attribute returns None, not AttributeError. Call .as_dict() on any object to get a plain dict. Reference fields (status, priority, assignee, queue, type…) are Reference objects — access .id, .display, .key, or .login directly without a second request. ### Issue AttributeNoteskeystr — 'QUEUE-42'summarystrdescriptionstr | NonestatusReference → .id e.g. 'inProgress', .display localized namepriorityReference → .id e.g. 'normal' / 'critical'typeReference → .id e.g. 'bug' / 'task'queueReference → .key e.g. 'QUEUE'assigneeReference | None → .login, .displayreporterReference → .login, .displaycreatedByReference → .logincreatedAtstr ISO-8601updatedAtstr ISO-8601deadlinestr | None — date '2026-03-01'tagslist[str]followerslist[Reference] → each .logincomponentslist[Reference] → each .displayfixVersionslist[Reference] → each .displaysprintlist[Reference] | None → each .displayparentReference | None → .keyvotesint ### Comment AttributeNotesidinttextstrtextHtmlstrcreatedByReference → .login, .displaycreatedAt / updatedAtstr ISO-8601summoneeslist[Reference]attachmentslist[Reference] ### Link AttributeNotesidinttypeReference → .id e.g. 'relates', 'blocks', 'is blocked by'directionstr — 'inward' / 'outward'objectReference → .key, .display (the linked issue)createdByReference → .logincreatedAtstr ISO-8601 ### Attachment AttributeNotesidintnamestr — filenamecontentstr — download URLmimetypestrsizeint — bytescreatedByReference → .logincreatedAtstr ISO-8601 ### Transition AttributeNotesidstr — transition key, e.g. 'close', 'start_progress'toReference → .id, .display (target status)screenReference | None ### Worklog entry AttributeNotesidintissueReference → .keycommentstr | Nonestartstr ISO-8601 datetimedurationstr ISO-8601 duration e.g. 'PT1H30M'createdByReference → .logincreatedAtstr ISO-8601 ### Queue AttributeNotesidintkeystrnamestrdescriptionstr | NoneleadReference → .login, .displayassignAutobooldefaultTypeReferencedefaultPriorityReferenceteamUserslist[Reference] → each .login ### User AttributeNotesuidintloginstrfirstName / lastNamestrdisplaystr — full nameemailstr ### BulkChange AttributeNotesidstrstatusstr — 'COMPLETE' / 'FAILED' / 'PROCESSING'statusTextstrexecutionChunkPercentintexecutionIssuePercentint ### Users # Find user by login, email, or display name — use when user says "assign to Иванов" for u in client.users.get_all(): print(u.login, u.display, u.email) # Get current user me = client.myself print(me.login) ### Sprints # Boards list for b in client.boards.get_all(): print(b.id, b.name) # Sprints for a board for s in client.boards[123].sprints.get_all(): print(s.id, s.name, s.status) # status: 'active','closed','draft' # Assign issue to sprint by id (found above) issue.update(sprint={'id': 456}) ### Error handling from yandex_tracker_client import exceptions try: issue = client.issues['QUEUE-99999'] except exceptions.NotFound: print('Issue not found') except exceptions.Forbidden: print('No access to this queue or issue') except exceptions.BadRequest as e: print('Invalid field or value:', e) except exceptions.Conflict: # Concurrent modification — re-fetch and retry issue = client.issues['QUEUE-42'] issue.update(...) Available exception classes: NotFound, Forbidden, BadRequest, Conflict, TrackerClientError (base). ### Notes org_id must be an int; cloud_org_id is a string. For Yandex Cloud orgs: use cloud_org_id= instead of org_id=, and optionally iam_token= for temporary IAM tokens instead of token=. To get valid resolution IDs for a queue: [r.id for r in client.resolutions.get_all()]. Print results clearly — use formatted strings, tables, or JSON so the user gets a readable summary. ## Trust - Source: tencent - Verification: Indexed source record - Publisher: Kandler3 - Version: 1.0.1 ## Source health - Status: healthy - Source download looks usable. - Yavira can redirect you to the upstream package for this source. - Health scope: source - Reason: direct_download_ok - Checked at: 2026-05-07T17:22:31.273Z - Expires at: 2026-05-14T17:22:31.273Z - Recommended action: Download for OpenClaw ## Links - [Detail page](https://openagent3.xyz/skills/yandex-tracker) - [Send to Agent page](https://openagent3.xyz/skills/yandex-tracker/agent) - [JSON manifest](https://openagent3.xyz/skills/yandex-tracker/agent.json) - [Markdown brief](https://openagent3.xyz/skills/yandex-tracker/agent.md) - [Download page](https://openagent3.xyz/downloads/yandex-tracker)