mission-report
Skill3 starsChanged 2 months agoMITView on GitHub
Write a Voltron Mission Report recording what each Lion shipped, blocked on, or deferred, as Markdown plus a schema.org JSON-LD knowledge graph. Use when the user asks to "write the mission report", "summarize the mission", "what shipped", or after Lions report back at the end of the Voltron mission flow, and whenever a report must be exported as structured or machine-readable output.
In this repository4 other files it gives its agents.
- browser-actionsSkill
- mission-briefSkill
- mission-planSkill
- skills-registrySkill
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 2 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 26
Load the plan's graph: Read mission-plan.jsonld and reuse every @id — mission, assignments, Lions, files.
- Very specific: line 47
Every FailedActionStatus needs an error a reader can act on.
- Distinctive: line 48
Do not quietly drop an assignment that was never dispatched.
- Distinctive: line 50
Keep the Markdown short. The detail lives in the graph.
≈ “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 Report — Markdown + Knowledge Graph
Emits the third and final Voltron mission document: the outcome of every assignment, recorded against the same graph nodes the plan created.
Read the format reference first: ${CLAUDE_PLUGIN_ROOT}/kb/mission-knowledge-format.md.
The report's value depends entirely on reusing the plan's identifiers — a report that mints
fresh @ids records nothing, because nothing links back to what was promised.
What a report contains
| Section | Content | Graph effect |
| Outcome | One line per assignment: shipped / blocked / deferred | Action.actionStatus transition |
| Mission Log | Each Lion's report verbatim-ish, in dispatch order | Action.result / Action.error |
| Success Criteria | Each brief criterion marked met or unmet, with evidence | DefinedTerm + evidence |
| Deferred | What was cut, and why | Action left Potential, with a reason |
Procedure
- Load the plan's graph. Read
mission-plan.jsonldand reuse every@id— mission, assignments, Lions, files. The report does not create assignments; it resolves them. - Transition each assignment's status using the KB's
actionStatusvocabulary: Lion reporteddone→CompletedActionStatus; Lion reportedblocked→FailedActionStatusplus anerrorstring naming the blocker; never dispatched → left atPotentialActionStatuswith aresultexplaining the deferral. - Record results honestly.
resultcarries what the Lion actually produced. If a Lion reporteddonebut its acceptance criterion was never checked, that is notCompletedActionStatus— mark itActiveand say the verification is outstanding. - Check success criteria against evidence, not against assignment status. An assignment can complete while its criterion still fails. Cite the evidence — a command's output, a test result, a file that now exists — for each criterion marked met.
- Set
endDateon the mission node. - Write
mission-report.md, thenmission-report.jsonld. Validate per the KB checklist and report both paths.
Writing rules
- Report what happened, not what was supposed to happen. A mission where two of five assignments blocked is a useful record; a report that rounds it to "mission complete" is not, and the graph will contradict it.
- Every
FailedActionStatusneeds anerrora reader can act on. "Failed" alone is noise. - Do not quietly drop an assignment that was never dispatched. Deferred work stays in the graph with its reason, or the next mission rediscovers it as new.
- Keep the Markdown short. The detail lives in the graph; the Markdown is the summary someone reads in thirty seconds.
What the accumulated graph answers
Because Lion and file identifiers are global across missions (see the KB's identifier scheme), a directory of mission reports answers questions no single report can: which files block most often, which Lion's assignments most frequently need a second pass, which success criteria keep recurring unmet. Mention this to the user once they have more than one mission on disk — it is the reason the format exists.
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.

