Install
$ agentstack add skill-marker-io-mcp-skills-marker-bug-to-fix ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
marker-bug-to-fix
Take a single Marker.io issue and turn it into a ready-to-review code change in the current repo. Cuts the report-to-fix round-trip from a manual context hunt to a minimal diff in minutes.
When to use
The user points at one Marker.io issue and wants it fixed or root-caused in this codebase: "fix marker issue <id>", "what's causing this bug", "investigate <marker link>". For triaging many issues at once, use marker-triage.
Steps
- Load the issue:
issue_get(full) andissue_get_context(browser/OS/screen + console + network
summaries). Read the reporter's description and the page URL.
- Pull the evidence that localizes the bug:
issue_get_screenshotfor the visual state, the failing
issue_get_console_log(logIndex) entries (stack traces), the failing/slow issue_get_network_request(requestIndex) (status, payload, response), and issue_get_attachments if any.
- Map signals to code: from the URL/route, error messages, stack frames, and failing endpoints, use
Grep/Glob/Read to find the responsible component, handler, or query in the working repo.
- State the root cause in one or two sentences. If you cannot localize it, say what extra info is
needed (and consider posting a gated comment_create asking the reporter).
- Propose a minimal fix as a diff. Prefer the smallest change that addresses the root cause. Show it
for review before changing files.
- Verify with the project's own tooling: hand off to the
/verifyskill, or run the relevant test,
to confirm the fix behaves.
- Open the PR via the
/prskill.
Marker writes (gated)
Propose, then apply after confirmation:
- Set the issue to an in-progress status (
issue_update_status, using a status discovered via
project_get).
- Post a root-cause + fix summary as a
comment_create(link the PR). Keep itMember-onlyunless the
reporter should see it.
Operating rules
This skill drives the Marker.io MCP. Read tools: projects_list, project_get, project_list_users, list_issue_types, issues_list, issue_get, issue_get_context, issue_get_screenshot, issue_get_attachments, issue_get_console_log, issue_get_network_request. Write tools: issue_update_status, issue_update_assignee, issue_update_priority, issue_update_type, comment_create.
- Resolve the project first. Call
projects_list(filter withsearchText). If more than one
could match, show the candidates and ask which one. Never guess a projectId.
- Discover before you write. Status and issue-type strings are project-specific. Call
list_issue_types and project_get for valid values before any issue_update_*. Get user IDs from project_list_users before assigning or @mentioning.
- Read cheap, drill deep only when it matters. Start from
issue_get_contextsummaries; fetch a
specific issue_get_console_log(logIndex) / issue_get_network_request(requestIndex) or issue_get_screenshot only when it changes a decision.
- Propose, then apply. Before any write, print a table of the planned changes (issue, field,
old to new). Apply via the *_update_* / comment_create tools only after the user confirms. Default comment visibility to Member-only unless the reporter should see it.
- Respect the limits.
issues_listdoes not return assignee (fetch per-issue viaissue_get).
The MCP cannot create Marker.io issues. Never embed tokens; rely on the configured MCP auth.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: marker-io
- Source: marker-io/mcp-skills
- License: MIT
- Homepage: https://marker.io
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.