Update to Agiflow
Snapshot Oct 24, 2026 · 12:56 UTC · version 3.0.1
Collection source: downloaded plugin package. These snapshots do not have a confirmed matching collection source. Differences in file lists alone do not establish changes to the package.
Payment or plan references changed
Skill instructions — Now says: Project Plan Turn a goal into a small set of well-scoped tasks in the project's Planning status. Planning tasks are not ready for execution;...
Observed in published text. Live prices and checkout terms have not been verified by this change.
Up to 10 plugins free · no card required.
All fields and original values.
Skill instructions
Sep 30, 2026 · 22:51 UTC
Agiflow Project Plan Create a focused near-term plan. Plan tasks first and leave work-unit creation to backlog grooming. ## Workflow 1. Call `get_current_scope` and resolve the target project with `list_projects` when needed. 2. Call `ge...
Oct 24, 2026 · 12:56 UTC
Project Plan Turn a goal into a small set of well-scoped tasks in the project's Planning status. Planning tasks are not ready for execution; `backlog-grooming` promotes them to Todo. ## Guardrails - Create tasks only in the Planning stat...
Compare saved observations
Download comparison JSONSKILL.md text changes
− removed · + added
--- before +++ after @@ -3,36 +3,73 @@ description: Turn a product goal or feature request into a clear Agiflow project plan with small, testable tasks in Planning status. Use when starting a project, decomposing a feature, clarifying requirements, or converting an idea into an actionable backlog. --- -# Agiflow Project Plan +# Project Plan -Create a focused near-term plan. Plan tasks first and leave work-unit creation to backlog grooming. +Turn a goal into a small set of well-scoped tasks in the project's Planning status. Planning tasks are not ready for execution; `backlog-grooming` promotes them to Todo. -## Workflow +## Guardrails -1. Call `get_current_scope` and resolve the target project with `list_projects` when needed. -2. Call `get_project_detail`, `list_project_statuses`, and `list_tasks` to understand goals, configured statuses, and existing work. -3. Ask only the questions needed to resolve the intended outcome, users, scope, constraints, and definition of done. -4. Check for duplicate or overlapping tasks before proposing new ones. -5. Decompose the outcome into small vertical slices that can be completed and verified independently. -6. For each proposed task, include: - - Clear title - - Outcome-focused description - - At least two testable acceptance criteria - - Explicit in-scope and out-of-scope boundaries - - Priority and useful categorization tags - - Dependencies or sequencing notes when relevant -7. Present the complete task plan and request approval before writing. -8. After approval, use `batch_create_tasks` for multiple tasks or `create_task` for one task. Use the exact Planning status name returned by `list_project_statuses`. -9. Call `list_tasks` to verify the created tasks and report any per-item errors. - -## Planning Rules - -- Do not create work units in this workflow. -- Do not promote tasks to Todo. -- Prefer the smallest plan that delivers useful progress. -- Split tasks that combine unrelated outcomes or cannot be verified independently. -- Preserve existing work and avoid duplicates. +- Create tasks only in the Planning status. Never create them in Todo. +- Do not create work units here. Grouping belongs to `backlog-grooming`. +- Ask follow-up questions about vague scope before creating anything. +- Present the plan and get explicit confirmation before calling any create tool. +- Favor the simplest version that delivers user-visible progress. Plan the next two to four weeks, not months. -## Response +## 1. Research first -Report created task titles, priorities, dependencies, and any tasks that still need user decisions. Recommend `refine-task` for ambiguous tasks or `backlog-grooming` when the Planning backlog is ready. +1. Resolve the project with `list_projects` and read it with `get_project_detail`. If the user needs a new project, confirm the name and details, then use `create_project`. +2. Call `list_project_statuses` to find the exact Planning status name. +3. Check for duplicates and related work with `list_tasks` and `list_work_units`. +4. Call `list_members` to see who could be assigned. + +Note whether the project is new (focus on foundations and setup) or existing (focus on integration with current work and avoiding overlap). + +## 2. Clarify requirements + +Extract and confirm: + +- **Who benefits and what they need** +- **In scope and explicitly out of scope** +- **Constraints**: deadlines, security, compatibility, performance +- **Dependencies**: other features, external services, data +- **How each outcome will be verified** + +If requirements are already clear, move on. If unknowns block confident planning, propose a time-boxed investigation task first. + +## 3. Decompose + +For fewer than five obvious tasks, skip to step 4. Otherwise: + +- Cut **vertical slices**: each delivers something demonstrable, not "all database work first". +- For each slice state: what it is (one sentence), what it needs, what it produces, and "done when" (WHEN action, THEN observable result). +- Identify the order (what must come first) and the smallest MVP subset. + +Present the decomposition and wait for confirmation. Do not create tasks yet. + +## 4. Create tasks in Planning + +After confirmation, use `batch_create_tasks` (or `create_task` for one) with: + +- **status**: the project's Planning status +- **title**: verb-led and specific ("Add", "Fix", "Update") +- **description**: context, constraints, and how it fits the goal; mention prerequisite tasks by title +- **acceptanceCriteria**: two to five concrete, verifiable conditions +- **priority**: low, medium, or high based on impact +- **tags**: consistent categories such as `feature:billing`, `bug:checkout`, `spike:search` +- **assignee**: optional; can be set during grooming + +Keep tasks small enough to finish in a short session. Include verification tasks where a criterion needs dedicated testing. Create prerequisite tasks first. + +## 5. Verify and summarize + +1. Call `list_tasks` filtered to the Planning status to confirm every task was created as intended. +2. Summarize: number of tasks created, grouping by tag, suggested order, MVP versus full scope, open questions or risks. +3. Recommend next steps: `refine-task` for any task that needs more depth, then `backlog-grooming` to prioritize, group, and promote. + +## Common mistakes + +- Creating tasks in Todo instead of Planning +- Creating work units during planning +- Horizontal slices ("all APIs first") +- Twenty tiny tasks when five would do, or two giant ones +- Acceptance criteria such as "works correctly"
Full technical diff · 1 changed fields
changed /skill_md_contents
"---\nname: project-plan\ndescription: Turn a product goal or feature request into a clear Agiflow project plan with small, testable tasks in Planning status. Use when starting a project, decomposing a feature, clarifying requirements, or converting an idea into an actionable backlog.\n---\n\n# Agiflow Project Plan\n\nCreate a focused near-term plan. Plan tasks first and leave work-unit creation to backlog grooming.\n\n## Workflow\n\n1. Call `get_current_scope` and resolve the target project with `list_projects` when needed.\n2. Call `get_project_detail`, `list_project_statuses`, and `list_tasks` to understand goals, configured statuses, and existing work.\n3. Ask only the questions needed to resolve the intended outcome, users, scope, constraints, and definition of done.\n4. Check for duplicate or overlapping tasks before proposing new ones.\n5. Decompose the outcome into small vertical slices that can be completed and verified independently.\n6. For each proposed task, include:\n - Clear title\n - Outcome-focused description\n - At least two testable acceptance criteria\n - Explicit in-scope and out-of-scope boundaries\n - Priority and useful categorization tags\n - Dependencies or sequencing notes when relevant\n7. Present the complete task plan and request approval before writing.\n8. After approval, use `batch_create_tasks` for multiple tasks or `create_task` for one task. Use the exact Planning status name returned by `list_project_statuses`.\n9. Call `list_tasks` to verify the created tasks and report any per-item errors.\n\n## Planning Rules\n\n- Do not create work units in this workflow.\n- Do not promote tasks to Todo.\n- Prefer the smallest plan that delivers useful progress.\n- Split tasks that combine unrelated outcomes or cannot be verified independently.\n- Preserve existing work and avoid duplicates.\n\n## Response\n\nReport created task titles, priorities, dependencies, and any tasks that still need user decisions. Recommend `refine-task` for ambiguous tasks or `backlog-grooming` when the Planning backlog is ready.\n"
"---\nname: project-plan\ndescription: Turn a product goal or feature request into a clear Agiflow project plan with small, testable tasks in Planning status. Use when starting a project, decomposing a feature, clarifying requirements, or converting an idea into an actionable backlog.\n---\n\n# Project Plan\n\nTurn a goal into a small set of well-scoped tasks in the project's Planning status. Planning tasks are not ready for execution; `backlog-grooming` promotes them to Todo.\n\n## Guardrails\n\n- Create tasks only in the Planning status. Never create them in Todo.\n- Do not create work units here. Grouping belongs to `backlog-grooming`.\n- Ask follow-up questions about vague scope before creating anything.\n- Present the plan and get explicit confirmation before calling any create tool.\n- Favor the simplest version that delivers user-visible progress. Plan the next two to four weeks, not months.\n\n## 1. Research first\n\n1. Resolve the project with `list_projects` and read it with `get_project_detail`. If the user needs a new project, confirm the name and details, then use `create_project`.\n2. Call `list_project_statuses` to find the exact Planning status name.\n3. Check for duplicates and related work with `list_tasks` and `list_work_units`.\n4. Call `list_members` to see who could be assigned.\n\nNote whether the project is new (focus on foundations and setup) or existing (focus on integration with current work and avoiding overlap).\n\n## 2. Clarify requirements\n\nExtract and confirm:\n\n- **Who benefits and what they need**\n- **In scope and explicitly out of scope**\n- **Constraints**: deadlines, security, compatibility, performance\n- **Dependencies**: other features, external services, data\n- **How each outcome will be verified**\n\nIf requirements are already clear, move on. If unknowns block confident planning, propose a time-boxed investigation task first.\n\n## 3. Decompose\n\nFor fewer than five obvious tasks, skip to step 4. Otherwise:\n\n- Cut **vertical slices**: each delivers something demonstrable, not \"all database work first\".\n- For each slice state: what it is (one sentence), what it needs, what it produces, and \"done when\" (WHEN action, THEN observable result).\n- Identify the order (what must come first) and the smallest MVP subset.\n\nPresent the decomposition and wait for confirmation. Do not create tasks yet.\n\n## 4. Create tasks in Planning\n\nAfter confirmation, use `batch_create_tasks` (or `create_task` for one) with:\n\n- **status**: the project's Planning status\n- **title**: verb-led and specific (\"Add\", \"Fix\", \"Update\")\n- **description**: context, constraints, and how it fits the goal; mention prerequisite tasks by title\n- **acceptanceCriteria**: two to five concrete, verifiable conditions\n- **priority**: low, medium, or high based on impact\n- **tags**: consistent categories such as `feature:billing`, `bug:checkout`, `spike:search`\n- **assignee**: optional; can be set during grooming\n\nKeep tasks small enough to finish in a short session. Include verification tasks where a criterion needs dedicated testing. Create prerequisite tasks first.\n\n## 5. Verify and summarize\n\n1. Call `list_tasks` filtered to the Planning status to confirm every task was created as intended.\n2. Summarize: number of tasks created, grouping by tag, suggested order, MVP versus full scope, open questions or risks.\n3. Recommend next steps: `refine-task` for any task that needs more depth, then `backlog-grooming` to prioritize, group, and promote.\n\n## Common mistakes\n\n- Creating tasks in Todo instead of Planning\n- Creating work units during planning\n- Horizontal slices (\"all APIs first\")\n- Twenty tiny tasks when five would do, or two giant ones\n- Acceptance criteria such as \"works correctly\"\n"
Full snapshot data
{
"description": "Turn a product goal or feature request into a clear Agiflow project plan with small, testable tasks in Planning status. Use when starting a project, decomposing a feature, clarifying requirements, or converting an idea into an actionable backlog.",
"included_files": [],
"name": "project-plan",
"skill_md_contents": "---\nname: project-plan\ndescription: Turn a product goal or feature request into a clear Agiflow project plan with small, testable tasks in Planning status. Use when starting a project, decomposing a feature, clarifying requirements, or converting an idea into an actionable backlog.\n---\n\n# Project Plan\n\nTurn a goal into a small set of well-scoped tasks in the project's Planning status. Planning tasks are not ready for execution; `backlog-grooming` promotes them to Todo.\n\n## Guardrails\n\n- Create tasks only in the Planning status. Never create them in Todo.\n- Do not create work units here. Grouping belongs to `backlog-grooming`.\n- Ask follow-up questions about vague scope before creating anything.\n- Present the plan and get explicit confirmation before calling any create tool.\n- Favor the simplest version that delivers user-visible progress. Plan the next two to four weeks, not months.\n\n## 1. Research first\n\n1. Resolve the project with `list_projects` and read it with `get_project_detail`. If the user needs a new project, confirm the name and details, then use `create_project`.\n2. Call `list_project_statuses` to find the exact Planning status name.\n3. Check for duplicates and related work with `list_tasks` and `list_work_units`.\n4. Call `list_members` to see who could be assigned.\n\nNote whether the project is new (focus on foundations and setup) or existing (focus on integration with current work and avoiding overlap).\n\n## 2. Clarify requirements\n\nExtract and confirm:\n\n- **Who benefits and what they need**\n- **In scope and explicitly out of scope**\n- **Constraints**: deadlines, security, compatibility, performance\n- **Dependencies**: other features, external services, data\n- **How each outcome will be verified**\n\nIf requirements are already clear, move on. If unknowns block confident planning, propose a time-boxed investigation task first.\n\n## 3. Decompose\n\nFor fewer than five obvious tasks, skip to step 4. Otherwise:\n\n- Cut **vertical slices**: each delivers something demonstrable, not \"all database work first\".\n- For each slice state: what it is (one sentence), what it needs, what it produces, and \"done when\" (WHEN action, THEN observable result).\n- Identify the order (what must come first) and the smallest MVP subset.\n\nPresent the decomposition and wait for confirmation. Do not create tasks yet.\n\n## 4. Create tasks in Planning\n\nAfter confirmation, use `batch_create_tasks` (or `create_task` for one) with:\n\n- **status**: the project's Planning status\n- **title**: verb-led and specific (\"Add\", \"Fix\", \"Update\")\n- **description**: context, constraints, and how it fits the goal; mention prerequisite tasks by title\n- **acceptanceCriteria**: two to five concrete, verifiable conditions\n- **priority**: low, medium, or high based on impact\n- **tags**: consistent categories such as `feature:billing`, `bug:checkout`, `spike:search`\n- **assignee**: optional; can be set during grooming\n\nKeep tasks small enough to finish in a short session. Include verification tasks where a criterion needs dedicated testing. Create prerequisite tasks first.\n\n## 5. Verify and summarize\n\n1. Call `list_tasks` filtered to the Planning status to confirm every task was created as intended.\n2. Summarize: number of tasks created, grouping by tag, suggested order, MVP versus full scope, open questions or risks.\n3. Recommend next steps: `refine-task` for any task that needs more depth, then `backlog-grooming` to prioritize, group, and promote.\n\n## Common mistakes\n\n- Creating tasks in Todo instead of Planning\n- Creating work units during planning\n- Horizontal slices (\"all APIs first\")\n- Twenty tiny tasks when five would do, or two giant ones\n- Acceptance criteria such as \"works correctly\"\n"
}SHA-256 of public snapshot: e18b9f7b86542d6c1506294cf8ad1793fea76169f47f3bb19e8b2217838bfcc3