AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

Report Skills

skill-dashaworks-report-skills-report-skills · by dashaworks

Shared setup and rules for the report-skills bundle (report-publisher, client-reporter, proposal-tracker, deck-publisher, report-designer). Loaded by the individual skills for the common ReportRoom MCP publish flow, account bootstrap, and safety gates. Not invoked directly — pick the specific skill for the job.

No reviews yet
0 installs
39 views
0.0% view→install

Install

$ agentstack add skill-dashaworks-report-skills-report-skills

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No 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.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-dashaworks-report-skills-report-skills)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
2mo ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

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 →
Are you the author of Report Skills? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Report Skills — shared flow and rules

All five skills publish through the ReportRoom MCP server. This file holds what they share; each skill's SKILL.md holds only what differs.

The publish flow (every skill follows this)

  1. Check access. Call account_status. If no account/API key exists, call create_account — it issues a key in-flow; never ask the user to leave the conversation to sign up. Note: publishing is blocked until the user clicks the email-verification link (create_account sends it), and unverified accounts serve from a preview/noindex domain — tell the user this upfront so the first publish isn't a surprise.
  2. Fetch the design language. Call get_design_system and follow it exactly when authoring. The design system supplies the taste; you supply content and assembly. Do not improvise your own CSS framework.
  3. Author a self-contained HTML document. Semantic HTML, responsive, dark-mode aware, no external dependencies beyond what the design system allows.
  4. Lint before publish. Call lint_document and fix every issue it reports. Never skip this step.
  5. Approval gate. Show the user: title, a one-paragraph summary of what the page contains, section list, and who it's for. Wait for explicit approval before calling publish on anything client-facing. For personal/internal shares the user may pre-approve ("just publish it").
  6. Publish. Call publish. Return the live URL prominently.
  7. Close the loop. Tell the user view analytics are attached: they can ask any time "how is my report doing?" (→ get_analytics, which returns 7-day views by day plus a summary). For tracked deliverables, offer to check back. Do not promise per-viewer or per-section detail — current analytics are aggregate view counts.

Safety rules (all skills, non-negotiable)

  • Published pages are public URLs. Before publishing, scan the content for secrets, API keys, internal hostnames, personal data of third parties, or anything marked confidential. If found, stop and ask.
  • Never fabricate data, metrics, or client results in a report. If a number is estimated, label it.
  • Never publish without the approval gate for client-facing content, even if the user seems in a hurry.
  • If lint_document or publish fails, show the actual error and fix the document — do not silently retry with degraded content.

Which skill for which job

| The user wants | Use | |---|---| | Share research/analysis they just produced | report-publisher | | A recurring, branded deliverable for a client | client-reporter | | A proposal/offer where knowing "did they open it" matters | proposal-tracker | | Slides, not a document | deck-publisher | | An existing page made better-looking | report-designer |

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.