AgentStack
Sign in Publish
MCP servers Skills Toolkits Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

Teamwork Research

skill-jinplu-teamwork-teamwork-research · by JinPLu

Use when the next safe step is to learn before acting: source-of-truth lookup, current API behavior, repro-surface framing, option comparison, stale-assumption refresh, or risk/evidence gathering before planning/debugging.

↗ Repository
— No reviews yet
0 installs
10 views
0.0% view→install

Install

$ agentstack add skill-jinplu-teamwork-teamwork-research

✓ 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.

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

About

Teamwork Research

Outcome

Establish project reality and return an evidence-backed answer, comparison, debug route, or plan-ready recommendation without implementing the change.

Enter When

Use when source of truth, current behavior, repro surface, alternatives, or risk is unclear. Treat a supplied article, URL, paper, repo, or report as seed evidence, not the research boundary. Route reproducible runtime failures to teamwork-debug; skip research only when direct evidence already resolves the uncertainty.

Do And Boundaries

Define the question and decision signal, check relevant prior artifacts, then split independent evidence tracks only when fan-out has clear value. Read local primary evidence first. Use current external sources when an API, dependency, model, platform, method, or upstream behavior may have changed; never send private source to public search. Distinguish important observed, inferred, and claimed findings, preserve meaningful dissent, and cite the evidence that supports the conclusion. Research does not authorize edits, external writes, or invented runtime values.

Write docs/teamwork/research/YYYY-MM-DD-.md only when findings need cross-turn reuse, support a durable plan or goal, or justify a consequential decision. External calibration alone does not require an artifact.

Done When

Return the question, evidence or verification, assumptions, options when relevant, recommendation, dissent or gaps, artifact path or none, and one route: teamwork-debug, teamwork-plan, follow-up research, or none.

Escalate

Ask for a user-owned decision when remaining uncertainty changes scope or risk; stop when required evidence or access is unavailable.

Conditional Protocols

Use research-protocol.md for lookup/deep research, subagent-dispatch.md and subagent-contract.md for fan-out, artifact-protocol.md for durable findings, and optional-skills.md for external tools. All paths are under skills/using-teamwork/references/.

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.

Write a review

Versions

  • v0.1.0 Imported from the upstream source.

More in AI & ML