agentleFS
kimgoetzkekimgoetzke/coding-agent-configs/skills/fetch-plan/SKILL.md

fetch-plan

Skill2 starsChanged 7 months agoMITView on GitHub

Retrieve saved plans, let the user choose one, then load its planning documents and summarise implementation status. Use when user asks to fetch, reopen, resume, continue, or inspect an existing plan.

The file

Show its rules and text

We split this skill into the rules it gives an agent

17 rules. About 0 are common, 12 are distinctive, and 5 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈12

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 5

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Use this skill to reopen an existing plan created with the planning-mode or planning skill.

    line 9
  2. Very specific:

    No output: use local-ai-folder-markdown, the default for this skill group

    line 19
  3. Very specific:

    Read ../planning/persistence/{implementation}.md and follow it whenever this skill lists or opens a plan, lists its documents or reads a …

    line 22
  4. Very specific:

    If no plans exist, tell the user no saved plans were found and suggest creating one with the planning-mode or planning skill.

    line 42
  5. Distinctive:

    If only a single plan exists, skip "Phase 2: Ask the user to choose" and consider this plan to be the one the user intends to select, then …

    line 44
  6. Distinctive:

    If the input is a number, resolve it against the numbered list

    line 53
  7. Distinctive:

    If the input is text, try exact plan-name match first

    line 54
  8. Distinctive:

    If no exact match exists, allow a single unambiguous case-insensitive partial match

    line 55
  9. Distinctive:

    If the input is invalid or ambiguous, explain why and ask again

    line 56
  10. Distinctive:

    Do not read any plan documents before a valid selection is resolved

    line 57
  11. Distinctive:

    After the user selects a plan, open it and read its available planning documents.

    line 61
  12. Distinctive:

    Read in this order:

    line 63
  13. Very specific:

    Only read documents that exist.

    line 70
  14. Distinctive:

    Prefer plain labels such as:

    line 82
  15. Distinctive:

    call out missing documents or uncertainty explicitly

    line 94
  16. Distinctive:

    After summarising status, prompt the user for the next step.

    line 99
  17. Distinctive:

    If the user chooses a follow-on action, continue from the selected plan rather than re-listing plans.

    line 105

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Fetch a Saved Plan

Use this skill to reopen an existing plan created with the planning-mode or planning skill.

Persistence

This skill belongs to the plans skill group. A persistence implementation decides where and how plans are stored. Before listing or opening plans, find the setting:

R="$(git rev-parse --show-toplevel 2>/dev/null)/.ai/config.properties"; { cat "$R" ~/.ai/config.properties | grep '^persistence\.plans[[:space:]]*=[[:space:]]*[^[:space:]]'; cat "$R" ~/.ai/config.properties | grep '^persistence[[:space:]]*=[[:space:]]*[^[:space:]]'; } 2>/dev/null | head -1 | tr -d ' \t\r'
  • No output: use local-ai-folder-markdown, the default for this skill group
  • persistence.plans={value} or persistence={value}: use {value}. If ../planning/persistence/{value}.md does not exist, tell the user the plans skill group has no {value} implementation and use local-ai-folder-markdown

Read ../planning/persistence/{implementation}.md and follow it whenever this skill lists or opens a plan, lists its documents or reads a planning document.

A selected plan may contain these planning documents:

  • plan
  • progress
  • findings
  • questions

Phase 1: List plans only

  1. List plans; do not read any planning documents yet
  2. Sort plans newest first
  3. Show the plan names exactly as stored, in a numbered list

Example:

  1. 2026-03-20 refactor-auth-flow
  2. 2026-03-18 add-retry-metrics

If no plans exist, tell the user no saved plans were found and suggest creating one with the planning-mode or planning skill.

If only a single plan exists, skip "Phase 2: Ask the user to choose" and consider this plan to be the one the user intends to select, then proceed to "Phase 3: Load the selected plan" and load this plan.

Phase 2: Ask the user to choose

Prompt the user to reply with either:

  • the list number, or
  • the plan name

Selection rules:

  • If the input is a number, resolve it against the numbered list
  • If the input is text, try exact plan-name match first
  • If no exact match exists, allow a single unambiguous case-insensitive partial match
  • If the input is invalid or ambiguous, explain why and ask again
  • Do not read any plan documents before a valid selection is resolved

Phase 3: Load the selected plan

After the user selects a plan, open it and read its available planning documents.

Read in this order:

  1. plan
  2. progress (if it exists)
  3. findings
  4. questions
  5. Any other documents in the plan, if relevant

Only read documents that exist. If the plan document is missing, say the plan is incomplete and continue with any other available docs.

Phase 4: Report implementation status

Summarise the plan's implementation status using only the loaded docs.

Base the status on:

  • plan: phase statuses and current phase
  • progress: latest actions and test results (only for multi-phase plans)
  • findings: blockers, issues, and decisions
  • questions: unresolved questions or missing user answers

Prefer plain labels such as:

  • Incomplete
  • Not started
  • In progress
  • Blocked
  • Complete
  • Unknown

When reporting back:

  • name the selected plan
  • state the best-fit implementation status
  • briefly justify it with concrete evidence from the docs
  • call out missing documents or uncertainty explicitly
  • mention any open questions or blockers

Phase 5: Ask how to proceed

After summarising status, prompt the user for the next step. Offer concise options such as:

  1. continue implementation
  2. review blockers and open questions
  3. improve or update the plan (only if incomplete)
  4. give a fuller summary of progress so far

If the user chooses a follow-on action, continue from the selected plan rather than re-listing plans.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.