mission-plan
Skill3 starsChanged 2 months agoMITView on GitHub
Write a Voltron Mission Plan (Lion Assignments with file ownership, Skills Gap analysis, Risk Register) as Markdown plus a schema.org JSON-LD knowledge graph. Use when the user asks to "write the mission plan", "draft assignments", "show the risk register", or at Steps 3-5 of the Voltron mission flow, and whenever a plan must be exported as structured or machine-readable output.
In this repository4 other files it gives its agents.
- browser-actionsSkill
- mission-briefSkill
- mission-reportSkill
- skills-registrySkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 48
If a Lion has no work this mission, omit them.
- Very specific: line 60
Do not dispatch Red, Blue, or Yellow Lion before that approval, and do not write mission-report until the Lions have actually reported.
≈ “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.
Files
Mission Plan — Markdown + Knowledge Graph
Emits the second of the three Voltron mission documents: what each Lion will do, what the mission cannot yet do, and what could go wrong.
Read the format reference first: ${CLAUDE_PLUGIN_ROOT}/kb/mission-knowledge-format.md.
It defines the JSON-LD envelope, the URN identifier scheme, the schema.org type mapping, the
actionStatus vocabulary, and the pre-write validation steps.
What a plan contains
| Section | Markdown form | Graph node |
| Lion Assignments | Table: Lion → Task → Files Owned → Depends On → Acceptance | Action per row |
| Skills Gap | Table: Capability → Status → Notes | DefinedTerm per row |
| Risk Register | Table: Category → Risk → Severity → Mitigation | Thing (additionalType: "Risk") per row |
Procedure
- Reuse the mission slug from the existing
mission-brief. Read the brief's.jsonldto recoverurn:voltron:mission:{slug}rather than re-deriving it from the objective — re-deriving drifts, and a drifted slug silently forks the mission into two graphs. - Draft assignments. One
Actionper Lion task. Enforce the file-ownership rule: no two assignments may list the same file inobject. This is checkable in the graph, so check it — group every assignment'sobject@ids and confirm no path appears twice. - Run Skills Gap via the
skills-registryskill and itsactions-map.json. Status iscovered,use-existing-skill:{name}, orneeds-new-skill. A registry entry markedconditionalcounts as covered only when itsrequiresprecondition is actually met — otherwise record the unmet precondition in Notes. - Build the Risk Register. Categories: technical, scope, integration, data-loss.
Severity is
L,M, orH. Every risk gets a mitigation, even if that mitigation is "accept and monitor". - Write
mission-plan.mdas three tables, thenmission-plan.jsonld. At plan stage every assignment carriesPotentialActionStatus— the plan describes intent, not outcome, and nothing isCompletedbefore the Lions have run. - Update the mission node so
memberlists every Lion that now has an assignment, andsubjectOfincludes the plan. - Validate per the KB checklist, then report both paths.
Writing rules
- Each assignment's Acceptance must be observable by someone who did not write the code.
- If a Lion has no work this mission, omit them. Do not pad the table to five rows.
Depends Onnames another assignment's@id, not a prose description — that is what makes the dispatch order derivable from the graph instead of from reading the table.- Files in
objectare repo-relative paths, matchingurn:voltron:file:{path}exactly.
The go/no-go gate
The plan is the document the user approves. After writing it, present the three tables and stop:
"Mission ready. Reply go to dispatch Lions, no-go to revise, or edit
<section>."
Do not dispatch Red, Blue, or Yellow Lion before that approval, and do not write
mission-report until the Lions have actually reported.
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
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.

