Workflow From Chats
- 1.2k installs
- 2.5k repo stars
- Updated August 5, 2026
- cursor/plugins
workflow-from-chats is a Cursor agent skill that mines recent chat transcripts for durable preferences and converts them into skills, rules, or workflow documentation for developers who want agent guidance learned from p
About
workflow-from-chats from cursor/plugins cursor-team-kit extracts reusable working preferences from recent Cursor chats instead of summarizing conversations. By default it scans the last seven days of parent transcripts plus relevant subagent evidence, hunting markers like I prefer, always, never, stop, review, PR, CI, and skill while redacting secrets and transcript paths. It clusters preference atoms by workflow shape—shipping, review, debugging, capture, communication—and chooses whether to emit a new skill, rule edit, workflow doc, or no artifact based on confidence ratings strong, medium, weak, or contradicted. Reach for workflow-from-chats when asked to learn preferences, mine feedback, personalize workflows, or generate person-specific or team-specific agent guidance from Cursor history.
- Extracts repeatable steps from chat sessions
- Packages workflows for Cursor plugins
- Standardizes agent runbooks and prompts
- Reduces manual re-prompting across tasks
- Improves team consistency for common fixes
Workflow From Chats by the numbers
- 1,240 all-time installs (skills.sh)
- +183 installs in the week ending Aug 4, 2026 (Skillselion tracking)
- Ranked #911 of 16,546 AI & Agent Building skills by installs in the Skillselion catalog
- Data as of Aug 5, 2026 (Skillselion catalog sync)
npx skills add https://github.com/cursor/plugins --skill workflow-from-chatsAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| Installs | 1.2k |
|---|---|
| repo stars | ★ 2.5k |
| Last updated | August 5, 2026 |
| Repository | cursor/plugins ↗ |
How do you turn Cursor chats into agent rules?
Turn Cursor chat threads into reusable agent workflows and plugin automations so repeated coding tasks run with consistent prompts, steps, and guardrails.
Who is it for?
Cursor developers who repeatedly correct agent behavior and want preferences codified into skills or rules without manually rewriting guidance from memory.
Skip if: One-off chat summaries, exporting full transcript archives, or teams without access to local Cursor parent conversation history.
When should I use this skill?
The user asks to learn preferences, mine feedback from recent chats, personalize workflows, or convert Cursor history into skills or rules.
What you get
Proposed skill, rule, or workflow doc drafts capturing durable triggers, decision rules, quality bars, and stop conditions extracted from chat evidence.
- skill draft
- rule draft
- workflow documentation
By the numbers
- Defaults to scanning the last 7 days of Cursor parent transcripts
- Defines 4 confidence ratings: strong, medium, weak, and contradicted
- Clusters preferences across 8 workflow shapes including shipping, review, and debugging
Files
Workflow From Chats
Infer durable working preferences from recent chats. Do not summarize chats; extract reusable workflow guidance.
Scope
- Default to the last 7 days unless the user asks for a different window.
- Read parent transcripts and relevant subagent transcripts. Use subagent content as evidence, but cite only parent conversations.
- Do not expose local transcript paths, secrets, customer data, private chat content, or credentials.
Workflow
1. State the target workflow or preference surface in one paragraph. 2. Build an internal transcript inventory: title/topic, parent conversation ID, approximate date, completion state, relevant subagents, and why it may contain preference evidence. 3. Scan for explicit preferences, corrections, and workflow markers such as "I prefer", "always", "never", "not what I asked", "stop", "review", "PR", "CI", "logs", and "skill". 4. Extract preference atoms: trigger, workflow step, decision rule, quality bar, stop condition, evidence, and confidence. 5. Rate confidence as strong, medium, weak, or contradicted. 6. Cluster by workflow shape rather than transcript: shipping, review, simplification, debugging, capture, communication, delegation, or validation. 7. Choose the artifact: new skill, skill edit, rule, workflow doc, or no artifact. 8. Draft only the reusable guidance. Filter anecdotes that will not help future tasks.
Confidence
- Strong: explicit user preference, workflow-changing correction, repeated parent-chat pattern, or direct request to encode behavior.
- Medium: accepted workflow, repeated tool/model/validation preference, or subagent consensus that the parent used successfully.
- Weak: agent-chosen behavior with no user feedback, one ambiguous transcript, or a likely task-specific correction.
- Contradicted: evidence points in incompatible directions; ask the user before writing files.
Artifact Choice
- Skill: recurring multi-step workflow with clear triggers.
- Rule: general behavior that should apply broadly.
- Workflow doc: useful context that is not reliably triggerable.
- No artifact: situational, stale, or low-confidence observation.
Output
Return a concise synthesis first:
- Target workflow.
- Evidence corpus with parent conversation citations only.
- Preference profile.
- Adopt, consider, dismissed.
- Proposed artifacts.
- Open questions only if they block writing.
Related skills
Setup Matt Pocock SkillsScaffold the per-repo configuration that Matt Pocock’s engineering agent skills rely on so they understand the issue tracker, triage labels, and domain documentation la495k196k
Lark AppsConnect Claude, Cursor or custom agents directly to Lark (Feishu) for messaging, document automation, approval workflows and enterprise data access.402k
CavemanSlash token usage by roughly 75% while keeping every technical detail intact when working with Claude Code, Cursor or similar agents.391k94.7k
Lark Skill MakerQuickly turn any Lark/Feishu OpenAPI call or multi-step workflow into a reusable agent skill with its own SKILL.md.386k16k
Running Claude Code Via Litellm CopilotRun Claude Code at a fraction of the cost by routing requests through LiteLLM to the GitHub Copilot Chat API.270k72
Codex PetGenerate a complete Codex Pet spritesheet and metadata from one reference image without needing an OpenAI key or Codex Pro.246k10How it compares
Use workflow-from-chats to learn habits from Cursor history; use create-skill when you already know the workflow and need a greenfield SKILL.md template.
FAQ
What time window does workflow-from-chats scan?
workflow-from-chats defaults to the last 7 days of Cursor parent transcripts unless the user requests a different range. It may read subagent transcripts as evidence but cites only parent conversations in outputs.
What artifacts can workflow-from-chats produce?
workflow-from-chats may propose a new skill for recurring multi-step workflows, a rule for broad behavior changes, a workflow doc for useful but non-triggerable context, or recommend no artifact when evidence is weak or contradictory.
Does workflow-from-chats summarize entire chats?
workflow-from-chats explicitly avoids summarizing chats. It extracts durable preference atoms such as triggers, decision rules, and stop conditions while filtering anecdotes, secrets, and low-confidence observations.